DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Applicant' s arguments, filed 4/14/2026, have been fully considered. The following rejections and/or objections are either reiterated or newly applied. They constitute the complete set presently being applied to the instant application.
Applicants have amended their claims, filed 4/14/2026, and therefore rejections newly made in the instant office action have been necessitated by amendment.
Claims 1-21 are the currently pending claims hereby under examination. Claims 1, 6, 8-11, 13, and 15-20 have been amended. Claim 21 is newly added.
Claim Interpretation
In claim 1, the phrase “discrete fluorescence data values … being discrete from fluorescence data values of each of the other frames” (lines 12-14) is interpreted consistent with the specification as requiring that each “frame” is its own grouped set of fluorescence samples (e.g., one set per rotation), not that the underlying imaged tissue region must be spatially non-overlapping between successive frames. The specification describes each “frame” as a distinct set of fluorescence data values collected for a given acquisition interval (e.g., one rotation / one B-scan), i.e., “the plurality of fluorescence data values is grouped into one or more frames … collected from one full rotation of the imaging catheter” (¶[0052]). Accordingly, “the discrete fluorescence data values of one frame … being discrete from fluorescence data values of each of the other frames” is reasonably interpreted as separate sets of fluorescence data values grouped per frame (i.e., distinct datasets per frame), without importing any requirement that successive frames must correspond to completely non-overlapping spatial regions (see also ¶[0072]–¶[0073]). This interpretation is consistent with that set forth in the prior Office Action (1/16/2026), except that the phrasing has been clarified to avoid any implication of temporal separation between frames (and adjusted to recite the amended claim language), as the claims and specification do not require such a limitation.
Similarly, claim 11 recites “discrete fluorescence data values … being discrete from fluorescence data values of each of the other frames” (lines 13-16). The same interpretation as found above in claim 1 also applies to claim 11.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-19 and 21 are rejected under 35 U.S.C. 103 as being unpatentable over Wang et al. (Wang, Hao et al. “Ex Vivo Catheter-Based Imaging of Coronary Atherosclerosis Using Multimodality OCT and NIRAF Excited at 633 Nm.” Biomedical optics express 6.4 (2015): 1363–1375. Web.), hereto referred as Wang 2015, and further in view of Ishihara (US-20120323072-A1), hereto referred as Ishihara, and further in view of Ughi et al. (Ughi, Giovanni J et al. “Clinical Characterization of Coronary Atherosclerosis With Dual-Modality OCT and Near-Infrared Autofluorescence Imaging.” JACC. Cardiovascular imaging 9.11 (2016): 1304–1314. Web.), hereto referred as Ughi 2016, and further in view of Ma et al. (US-20210116380-A1), hereto referred as Ma, and further in view of Lemire (Lemire, Daniel. “Streaming Maximum-Minimum Filter Using No More than Three Comparisons per Element.” ArXiv abs/cs/0610046 (2007)), hereto referred as Lemire.
Regarding claim 1, Wang 2015 teaches that a method for accurately quantifying fluorescence emitted from within a bodily lumen (Wang 2015, Abstract: “In this study, we used a recently developed imaging system and double-clad fiber (DCF) catheter capable of simultaneously acquiring both OCT and red excited near-infrared autofluorescence (NIRAF) images (excitation: 633 nm, emission: 680nm to 900nm)... We found that NIRAF is elevated in lesions that contain necrotic core – a feature that is critical for vulnerable plaque diagnosis... These results suggest that multimodality intracoronary OCT-NIRAF imaging technology may be used in the future to provide improved characterization of coronary artery disease in human patients”, this shows Wang 2015 teaches a catheter-based method of quantifying fluorescence emitted from within a bodily lumen for medical imaging purposes); the method comprising: scanning the bodily lumen... with an imaging catheter that transmits a light of wavelength capable of stimulating emission of fluorescence from within different regions of the bodily lumen (Wang 2015, p. 4, Sec. 2.1.1: “The multimode inner cladding (NA ≥ 0.46, diameter = 124–126 µm, fiber outer diameter = 250 µm) was used to guide 633 nm excitation light and to collect NIRAF tissue emission. The working distances for OCT and NIRAF were 2 mm and 0.5 mm, respectively. The focal spot size for OCT and NIRAF was 27 µm and 100 µm, respectively”, this teaches that the catheter transmits excitation light at 633 nm to stimulate fluorescence emission from within tissue regions of the bodily lumen); collecting a plurality of fluorescence data values corresponding to the fluorescence emitted from the different regions within the bodily lumen while the scanning is being performed... (Wang 2015, p. 7, Sec. 2.2.3: “The sampling rate of NIRAF signal was 40 kHz, enabling integration of the NIRAF emission within the period of a single A-line. Each OFDI frame consisted of 1,024 A-lines, with a frame rate of 39 frames/sec (40 kHz/1024). The rotary junction was operated at 39 rotations per second so each OFDI image covered one rotation”, which teaches that fluorescence data values are acquired in real-time during scanning at the A-line level and organized into frames corresponding to catheter rotation, thereby teaching collection of a plurality of fluorescence data values corresponding to emission from different regions of the bodily lumen during the scanning process); grouping the plurality of fluorescence data values into one or more frames, each frame of the one or more frames having a number of discrete fluorescence data values of the plurality of fluorescence data values collected from one full rotation of the imaging catheter within the bodily lumen (Wang 2015, p. 7, Sec. 2.2.3: “The sampling rate of NIRAF signal was 40 kHz, enabling integration of the NIRAF emission within the period of a single A-line. Each OFDI frame consisted of 1,024 A-lines, with a frame rate of 39 frames/sec (40 kHz/1024). The rotary junction was operated at 39 rotations per second so each OFDI image covered one rotation”, which teaches that each frame comprises individual A-line-based fluorescence measurements that are separately acquired within a single rotation, such that the fluorescence data values within a frame are discrete from one another and from fluorescence data values of other frames); and the discrete fluorescence data values of one frame of the one or more frames being discrete from fluorescence data values of each of the other frames of the one or more frames (Wang 2015, p. 7, Sec. 2.2.3: “Each OFDI frame consisted of 1,024 A-lines, with a frame rate of 39 frames/sec (40 kHz/1024). The rotary junction was operated at 39 rotations per second so each OFDI image covered one rotation”, this teaches that each frame is a distinct set of A-line based data acquired during a single rotation, such that the fluorescence data values of one frame are separate from the fluorescence data values of other frames corresponding to other rotations).
Also regarding claim 1, Wang 2015 does not fully teach performing the methods of scanning, collecting, and calculating while the scanning is being performed in the in vivo environment. Rather, Wang 2015 teaches imaging and fluorescence data acquisition using a catheter system, but the imaging is performed on ex vivo human cadaver coronary arteries (Wang 2015, p. 8, Sec. 2.2.6: “A total of 15 coronary arteries from 5 human cadaver hearts were imaged using the multimodality OCT-NIRAF system and catheter”). Wang 2015 further indicates that the disclosed imaging technique is intended for clinical application, stating that “These results suggest that multimodality intracoronary OCT-NIRAF imaging technology may be used in the future to provide improved characterization of coronary artery disease in human patients” (Wang 2015, Abstract), thereby indicating that the ex vivo imaging is performed as validation of a system intended for in vivo use. However, it does not explicitly teach scanning in an in vivo environment.
Ughi 2016 teaches that “The authors present the clinical imaging of human coronary arteries in vivo using a multimodality optical coherence tomography (OCT) and near-infrared autofluorescence (NIRAF) intravascular imaging system and catheter” (Ughi 2016, Abstract) and further teaches that “OCT-NIRAF imaging of the vessel undergoing PCI was performed in all cases” (Ughi 2016, p. 1306), thereby demonstrating catheter-based fluorescence imaging performed in vivo during active pullback scanning in human patients.
Ishihara teaches that fluorescence image processing is performed on a per-frame basis as successive frames are generated, including calculating an “average gradation value m (=1050) of the entire image” (Ishihara, ¶[0083]) and an “average gradation value m (=1365) of the entire image of the subsequent frame” (Ishihara, ¶[0088]), and further teaches that processed images are immediately displayed such that “a new corrected fluorescence image... is displayed on the monitor 50” (Ishihara, ¶[0084]). Because each frame corresponds to image data generated during ongoing acquisition, and Ishihara performs calculations and display updates upon generation of each successive frame, the processing is performed contemporaneously with acquisition of the frames, which in the context of catheter pullback as evidenced by Ughi 2016 corresponds to processing during scanning. Ishihara additionally teaches that an operator may adjust a threshold value during generation of corrected fluorescence images of subsequent frames, thereby evidencing that processing is performed during ongoing frame generation.
It would have been prima facie obvious before the effective filing date of the claimed invention to modify Wang 2015 in view of Ughi 2016 and Ishihara to perform the scanning of the bodily lumen in an in vivo environment and to perform the processing steps (collecting and calculating) during the scanning process because Ughi 2016 demonstrates that the same type of catheter-based OCT-NIRAF imaging system is used clinically in vivo, and Ishihara demonstrates that fluorescence image data is processed on a per-frame basis as frames are generated and displayed, including operator interaction during ongoing image generation, thereby evidencing real-time or near real-time processing of acquired image data. Adapting Wang’s catheter-based imaging technique for in vivo use and applying Ishihara’s per-frame processing during acquisition represents a predictable use of known imaging and data processing techniques to enable real-time or near real-time evaluation of fluorescence data during catheter-based imaging while maintaining the underlying catheter-based OCT-NIRAF imaging architecture of Wang 2015, including its excitation, detection, and frame-based acquisition processes. Accordingly, the modification does not alter the fundamental principle of operation of Wang 2015 but instead applies known processing techniques within its existing framework. The benefit of the combination is that performing fluorescence data processing during in vivo scanning enables real-time or near real-time evaluation of fluorescence signals, thereby improving clinical assessment of tissue characteristics during the procedure.
Also regarding claim 1, the modified Wang 2015 does not fully teach calculating a central tendency of the discrete fluorescence data values in each of the one or more frames while the scanning is being performed in the in vivo environment. Rather, the modified Wang 2015 teaches collecting fluorescence intensity data in rotation-synchronized frames and performing normalization of fluorescence data, but it does not teach calculating a central tendency, such as an average or mean, of the discrete fluorescence data values in each frame while the scanning is being performed in the in vivo environment.
Ishihara teaches calculating an “average gradation value m (=1050) of the entire image” (Ishihara, ¶[0083]) and further calculating an “average gradation value m (=1365) of the entire image of the subsequent frame” (Ishihara, ¶[0088]), thereby teaching a per-frame central tendency calculation performed sequentially as frames are acquired for each frame in the sequence. Because the gradation values of Ishihara correspond to pixel intensity values representing fluorescence emission at each pixel, and each image corresponds to a frame of such values, the calculated average directly corresponds to a central tendency of discrete fluorescence data values within a frame under the broadest reasonable interpretation.
It would have been prima facie obvious before the effective filing date of the claimed invention to have further modified the modified Wang 2015 in view of Ishihara to calculate a central tendency of the discrete fluorescence data values in each of the one or more frames while the scanning is being performed in the in vivo environment. It would have been prima facie obvious before the effective filing date of the claimed invention to compute a per-frame central tendency for the fluorescence data of Wang 2015 because frame-based statistical summarization of image intensity data is a well-known and predictable processing step used to reduce data dimensionality and enable subsequent frame-level comparisons and thresholding operations, and Ishihara explicitly demonstrates such frame-level statistical processing applied to sequential image frames. This modification would have been possible because Wang 2015 already forms rotation-based frames from discrete fluorescence measurements, Ishihara expressly performs a per-frame average calculation upon acquisition of each subsequent frame, and Ughi 2016 evidences that such catheter-based fluorescence imaging is performed in vivo during pullback scanning. The benefit of the combination is that a frame-level central tendency provides a consistent statistical basis for subsequent frame-level threshold and background processing while remaining within the existing catheter imaging pipeline. Further, because Ishihara performs the average calculation upon acquisition of each subsequent frame and Ughi 2016 demonstrates that such frames are acquired during active catheter pullback, the calculation of the central tendency necessarily occurs while the scanning is being performed in the in vivo environment.
Also regarding claim 1, the modified Wang 2015 does not fully teach assigning the central tendency of one of the one or more frames as a background fluorescence threshold for the in vivo environment by determining a lowest central tendency of the one or more frames via, until the scanning is finished, setting a first central tendency of a first frame of the one or more frames as a pending central tendency, iteratively comparing the respective central tendency of each of the following one or more frames to the pending central tendency, setting any respective central tendency that is lower than the pending central tendency as the pending central tendency, and once the scanning has finished assigning the pending central tendency as the lowest central tendency and as the background fluorescence threshold for the in vivo environment. Rather, the modified Wang 2015 teaches collecting fluorescence data in rotation-synchronized frames during catheter pullback, performing per-frame central tendency calculations as modified above in view of Ishihara, and using fluorescence intensity values as part of normalization processing in fluorescence imaging, but the modified Wang 2015 does not fully teach the specifically recited iterative pending-minimum process in which a first frame central tendency is set as a pending central tendency, each subsequent frame central tendency is iteratively compared to the pending central tendency during scanning, any lower central tendency replaces the pending central tendency, and the pending central tendency is assigned at the end of scanning as the lowest central tendency and background fluorescence threshold.
Ishihara teaches calculating per-frame average values for sequentially acquired frames (Ishihara, ¶[0083], ¶[0088]).
Ughi 2016 evidences that fluorescence intensity varies across frames during in vivo acquisition, including lower-intensity measurements (Ughi 2016, p. 1306). The gradation values processed by Ishihara correspond to pixel intensity values of an image, which are directly analogous to fluorescence intensity values in Wang 2015, as both represent scalar intensity measurements associated with image data. Specifically, Ishihara’s corrected fluorescence image is generated by dividing the fluorescence image by a white-light reference image (Ishihara, ¶[0046]), such that the gradation values in the corrected fluorescence image are quantitative representations of fluorescence emission intensity at each pixel, normalized for observation distance and angle. These gradation values represent the same physical quantity as the fluorescence data values of Wang 2015, expressed in digital units, and statistical processing applied to such values is equivalently applicable across both systems.
Ma teaches that “A background correction method for a fluorescence microscopy system includes receiving a raw image stack, determining a number of temporal minimum intensity values for each pixel location from the raw image stack, and calculating an expected background value for each pixel location based on the number of temporal minimum intensity values for the pixel location” (Ma, Abstract). Ma is in the analogous field of fluorescence image processing and background correction, which involves the same problem of distinguishing signal from background intensity in fluorescence imaging data as in Wang 2015. Ma further teaches “RECEIVE RAW IMAGE STACK", “SEGMENT THE IMAGE STACK INTO A NUMBER OF SUB-STACKS", “CALCULATE THE MINIMUM VALUE OF THE PIXELS ALONG THE TEMPORAL AXIS FOR EACH PIXEL LOCATION FOR EACH SUB-STACK", and “CALCULATE THE EXPECTED BACKGROUND VALUE FOR EACH PIXEL LOCATION ACCORDING TO A DERIVED TRANSFER FUNCTION BASED ON THE OBTAINED MINIMUM VALUES FOR EACH PIXEL LOCATION” (Ma, Fig. 3), thereby teaching determining minimum intensity values over time and using those values to calculate background intensity. A person of ordinary skill in the art would have recognized that Ma’s use of temporal minimum intensity values to estimate background conditions at the pixel level applies equivalently to frame-level statistical representations in a catheter pullback context, where each frame corresponds to a different spatial region and per-pixel temporal tracking is not feasible. It would have been obvious to adapt Ma’s minimum-based background determination to operate on per-frame central tendencies calculated by Ishihara by identifying the lowest per-frame central tendency across the pullback as the background fluorescence reference, which represents a predictable application of Ma’s minimum-based background estimation principle using a more statistically robust frame-level metric. The initialization step of setting the first central tendency as the pending central tendency is supported by Ishihara’s FIG. 7 process, in which the threshold-value setting unit acquires the first corrected fluorescence image, calculates the average gradation value m (=1050) of that initial frame, and stores it as the operative baseline value from which subsequent comparisons proceed (Ishihara, ¶[0083]).
Lemire teaches that computing a minimum over a sequence is performed via comparisons across the values of the sequence, stating that “Computing either the global maximum or the global minimum of an array of n elements requires n − 1 comparisons” (Lemire, p. 1), and further teaches that such running minimum computations can be performed with zero stream latency, defined as “the maximum number of data points required after the window has passed” (Lemire, Table I), thereby evidencing that minimum determination may be implemented through sequential comparison of incoming data values as they are acquired without requiring future data. This inherently corresponds to an iterative process in which an initial value is selected, each subsequent value is compared to the current minimum, the current minimum is updated when a lower value is identified, and the final minimum is obtained after processing all values in the sequence.
It would have been prima facie obvious before the effective filing date of the claimed invention to have further modified the modified Wang 2015 in view of Ishihara, Ughi 2016, Ma, and Lemire to assign the central tendency of one of the one or more frames as a background fluorescence threshold for the in vivo environment by determining a lowest central tendency using the claimed iterative comparison process. This modification would have been possible because the modified Wang 2015 already provides rotation-synchronized frame data acquired during catheter pullback, Ishihara already calculates a per-frame central tendency for each acquired frame, Ma teaches determining minimum intensity values over time and using those values to calculate background intensity (Ma, Abstract; Fig. 3), and Lemire teaches sequential comparison of values to determine a minimum over a sequence. Further, assigning the lowest value after scanning the completes is implicit because the determination of a minimum across a sequence of values requires processing the entire sequence, and the final minimum cannot be known until all values have been evaluated, wherein the catheter pullback defines a finite sequence of frames that must be fully processed before the lowest central tendency can be determined. The benefit of the combination is improved robustness in determining a background fluorescence threshold from in vivo sequential frame data by using an iterative lowest-value tracking process that identifies the lowest background-dominant frame statistic across the scan. The combination of Ishihara’s per-frame central tendency calculation, Ma’s use of minimum values across temporally acquired image data for background estimation, and Lemire’s sequential comparison methodology collectively yields the iterative process of initializing a pending value, comparing each subsequently computed frame-level value, updating the pending value when a lower value is identified, and finalizing the lowest value after completion of data acquisition, which represents a predictable integration of known data processing techniques.
Also regarding claim 1, the modified Wang 2015 does not fully teach adjusting the fluorescence data values of the one or more frames based on the background fluorescence threshold for the in vivo environment by subtracting a background fluorescence value corresponding to the background fluorescence threshold from each of the fluorescence data values or by removing any fluorescence data value from the fluorescence data values that is equal to or lower than the background fluorescence threshold, thereby generating adjusted fluorescence data values. Rather, the modified Wang 2015 teaches adjusting fluorescence data by subtracting a PBS-derived background signal and, as modified above, teaches determining a background fluorescence threshold from frame-based processing, but the modified Wang 2015 does not fully teach the specifically recited alternatives of adjusting the fluorescence data values based on the determined background fluorescence threshold by subtracting a corresponding background fluorescence value from each fluorescence data value or by removing any fluorescence data value from the fluorescence data values that is equal to or lower than the background fluorescence threshold.
Wang teaches subtracting background fluorescence using PBS-derived background values (p. 7–8, Sec. 2.2.5). Ishihara teaches threshold-based adjustment including replacing values below a threshold with zero (¶[0084]: “the image adjuster 51 replaces the gradation values of pixels having gradation values smaller than the threshold value S (=1441) with zero”; ¶[0100]: “the image adjuster 51 may adjust the gradation values of the corrected fluorescence image on the basis of the current threshold value S”).
It would have been prima facie obvious before the effective filing date of the claimed invention to have further modified the modified Wang 2015 in view of Ishihara to adjust the fluorescence data values of the one or more frames based on the background fluorescence threshold for the in vivo environment by subtracting a background fluorescence value corresponding to the background fluorescence threshold from each fluorescence data value or by removing any fluorescence data value from the fluorescence data values that is equal to or lower than the background fluorescence threshold, thereby generating adjusted fluorescence data values. This modification would have been possible because Wang already performs background subtraction from fluorescence data and Ishihara teaches threshold-based removal of lower-valued fluorescence image data. The benefit of the combination is improved suppression of background-level fluorescence contributions while retaining lesion-related fluorescence signal, thereby improving contrast and interpretability in the resulting fluorescence images.
Also regarding claim 1, the modified Wang 2015 does not fully teach generating an image based on the adjusted fluorescence data values. Rather, the modified Wang 2015 teaches generating fluorescence images from processed fluorescence data and, as modified above, teaches adjusting fluorescence data values based on a background fluorescence threshold, but the modified Wang 2015 does not fully teach generating an image specifically based on fluorescence data values adjusted using the claimed threshold-based processing.
Wang 2015 teaches that “The normalized NIRAF signal was rendered as a ring shaped image, where each NIRAF data point matches the corresponding OCT A-line at each given rotational position of the catheter” (Wang 2015, p. 7–8).
Ishihara teaches generating and displaying corrected fluorescence images after threshold-based adjustment (¶[0084]: “a new corrected fluorescence image, in which 91.8% of the display of the background is eliminated and 89.5% of the display of the lesion is maintained, is displayed on the monitor 50”; ¶[0100]: “the image adjuster 51 may adjust the gradation values of the corrected fluorescence image on the basis of the current threshold value S”).
It would have been prima facie obvious before the effective filing date of the claimed invention to have further modified the modified Wang 2015 in view of Ishihara to generate an image based on the adjusted fluorescence data values because Ishihara explicitly uses threshold-adjusted fluorescence image values to generate and display a corrected fluorescence image. The benefit of the combination is improved visual contrast and suppression of background fluorescence in the generated image while preserving lesion-related fluorescence signal.
Regarding claim 2, the modified Wang 2015 partially teaches generating the image includes generating an image of the bodily lumen based on fluorescence data values of the plurality of fluorescence data values that are larger than a value of the background fluorescence threshold. Specifically, the modified Wang 2015 renders NIRAF images after background subtraction and normalization, as shown in claim 1, but it does not specify that the rendered image is generated based on only those fluorescence data values that are larger than a background fluorescence threshold (Wang 2015, p. 7–8, Sec. 2.2.5; Wang 2015, p. 8–9, Sec. 2.2.5).
Ishihara teaches generating a corrected fluorescence image by suppressing values below a threshold value S, specifically teaching that “the image adjuster 51 replaces the gradation values of pixels having gradation values Smaller than the threshold value S (=1575) with zero”, which results in an image that is generated based on values larger than the threshold value because values lower than the threshold do not contribute to the displayed fluorescence signal (Ishihara, ¶[0065]). Ishihara further teaches displaying the resulting corrected fluorescence image on a monitor (Ishihara, ¶[0066]: “a new corrected fluorescence image with increased contrast between the area displaying the lesion and the area displaying the background is displayed on the monitor 50”).
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have further modified the modified Wang 2015 in view of Ishihara to generate an image of the bodily lumen based on fluorescence data values that are larger than a value of the background fluorescence threshold by applying a background fluorescence threshold to suppress fluorescence data values at or below the threshold (e.g., setting values below the threshold to zero) and generating the image from the remaining values above the threshold (Ishihara, ¶[0065]; Ishihara, ¶[0066]). This modification is feasible because the modified Wang 2015 already renders fluorescence images from per-rotation frame data after post-processing, and Ishihara’s threshold-based suppression is a software-side masking operation applied to the already-generated fluorescence values without changing catheter structure, rotation rate, pullback acquisition, or image rendering architecture. The benefit of the combination is improved image contrast and robustness by suppressing background-level fluorescence contributions and emphasizing true fluorescence signal associated with lesions during display.
Regarding claim 3, the modified Wang 2015 partially teaches generating the image includes generating an image of the bodily lumen based on fluorescence data values of the plurality of fluorescence data values that are larger than a value of the background fluorescence threshold and/or removing fluorescence data values of the plurality of fluorescence data values that are lower than or equal to the value of the background fluorescence threshold. Specifically, the modified Wang 2015 renders NIRAF images after background subtraction and normalization, as shown in claim 1, but it does not specify that the rendered image is generated based on only those fluorescence data values that are larger than a background fluorescence threshold and/or that fluorescence data values lower than or equal to the background fluorescence threshold are removed (Wang 2015, p. 7–8, Sec. 2.2.5; Wang 2015, p. 8–9, Sec. 2.2.5–2.2.6).
Ishihara teaches generating a corrected fluorescence image by suppressing values below a threshold value S, specifically teaching that “the image adjuster 51 replaces the gradation values of pixels having gradation values Smaller than the threshold value S (=1575) with zero”, which removes or suppresses values at or below the threshold and yields an image that is generated based on values larger than the threshold because values lower than the threshold do not contribute to the displayed fluorescence signal (Ishihara, ¶[0065]). Ishihara further teaches displaying the resulting corrected fluorescence image on a monitor (Ishihara, ¶[0066]: “a new corrected fluorescence image with increased contrast between the area displaying the lesion and the area displaying the background is displayed on the monitor 50”).
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have further modified the modified Wang 2015 in view of Ishihara to generate an image of the bodily lumen based on fluorescence data values that are larger than a value of the background fluorescence threshold and/or to remove fluorescence data values that are lower than or equal to the value of the background fluorescence threshold by applying a background fluorescence threshold to suppress fluorescence data values at or below the threshold (e.g., setting values below the threshold to zero) and generating the image from the remaining values above the threshold. This modification is feasible because the modified Wang 2015 already renders fluorescence images from per-rotation frame data after post-processing, and Ishihara’s threshold-based suppression is a software-side masking operation applied to the already-generated fluorescence values without changing catheter structure, rotation rate, pullback acquisition, or image rendering architecture. The benefit of the combination is improved image contrast and robustness by suppressing background-level fluorescence contributions and emphasizing true fluorescence signal associated with lesions during display.
Regarding claim 4, the modified Wang 2015 teaches that the one or more frames includes one B-scan frame (Wang 2015, p. 7, Sec. 2.2.3: “Each OFDI frame consisted of 1,024 A-lines, with a frame rate of 39 frames/sec (40 kHz/1024). The rotary junction was operated at 39 rotations per second so each OFDI image covered one rotation”, Wang uses the terms “frame” and “image” to denote a cross-sectional acquisition produced by one full rotation, which corresponds to a B-scan frame in the claim) the plurality of fluorescence data values respectively corresponds to a plurality of A-lines of the one B-scan frame collected by the imaging catheter in one full revolution within the bodily lumen (Wang 2015, p. 7–8, Sec. 2.2.5–2.2.6: “The normalized NIRAF signal was rendered as a ring shaped image, where each NIRAF data point matches the corresponding OCT A-line at each given rotational position of the catheter”, this shows that NIRAF fluorescence data points correspond one-to-one with OCT A-lines within a rotation-synchronized frame; Wang 2015, p. 7, Sec. 2.2.3: “The rotary junction was operated at 39 rotations per second so each OFDI image covered one rotation”, this shows that each frame is produced during one full revolution of the catheter within the lumen); and collecting the plurality of fluorescence data values includes collecting one fluorescence data value for each of the A-lines (Wang 2015, p. 7, Sec. 2.2.3: “The sampling rate of NIRAF signal was 40 kHz, enabling integration of the NIRAF emission within the period of a single A-line”, this shows one fluorescence measurement per A-line; Wang 2015, p. 7–8, Sec. 2.2.5–2.2.6: “each NIRAF data point matches the corresponding OCT A-line”, corroborating one data value per A-line)
Also regarding claim 4, the modified Wang 2015 does not fully teach calculating the central tendency includes calculating a mean of the fluorescence data values corresponding to all of the A-lines in the one B-scan frame. Rather, the modified Wang 2015 teaches rotation-synchronized frames composed of A-lines, acquisition of a fluorescence value per A-line, and one-to-one correspondence between fluorescence data values and A-lines within a single revolution, but does not explicitly teach calculating a mean of the fluorescence data values corresponding to all of the A-lines in a single frame (Wang 2015, p. 7, Sec. 2.2.3; p. 7–8, Sec. 2.2.5–2.2.6).
Ishihara teaches calculating a per-frame mean by teaching that an average gradation value m is calculated for an acquired frame and stored for subsequent processing (Ishihara, ¶[0083]), and further teaches calculating such average values for subsequent frames (Ishihara, ¶[0088]), thereby establishing that a central tendency (mean) is calculated for each individual frame based on the image data of that frame.
It would have been prima facie obvious before the effective filing date of the claimed invention to have further modified the modified Wang 2015 in view of Ishihara to calculate a mean of the fluorescence data values corresponding to all of the A-lines in a single rotation-synchronized frame by calculating, for each acquired frame, an average of the frame’s fluorescence image values as a frame-level mean. This modification would have been possible because Wang 2015 already acquires one fluorescence data value per A-line and groups those values into a per-rotation frame, and Ishihara teaches a software-side operation of calculating an average value for each frame based on the image data of that frame without modifying the underlying acquisition system. The benefit of the combination is improved robustness and consistency by providing a frame-level mean that supports subsequent processing and comparison across frames.
Regarding claim 5, as shown in claim 1, the modified Wang 2015 teaches that the one or more frames include a plurality of B-scan frames collected by the imaging catheter during an entire pullback within the bodily lumen (Wang 2015, p. 7, Sec. 2.2.3: “Each OFDI frame consisted of 1,024 A-lines, with a frame rate of 39 frames/sec (40 kHz/1024). The rotary junction was operated at 39 rotations per second so each OFDI image covered one rotation. The pullback rate was 5 mm/s”, which teaches that the rotation-synchronized frames established in claim 1 are acquired repeatedly at a frame rate while the catheter is pulled back through the lumen, thereby yielding a plurality of B-scan frames during pullback; p. 7, Sec. 2.2.3: “The sheath was inserted into the coronary artery and the starting point of the pullback was labeled with tissue marking ink… The endpoint of the pullback was also labeled by tissue marking ink”, which teaches that the imaging pullback has defined start and end points along the bodily lumen, consistent with collecting the rotation-synchronized frames of claim 1 throughout the pullback length; p. 7, Sec. 2.2.3: “Since the start and end points of the OCT-NIRAF pullback were marked on the coronary artery tissue before cutting, the position of each histology slide was matched with OCT-NIRAF frame”, which teaches that multiple OCT-NIRAF rotation-synchronized frames were acquired along the pullback such that the frames correspond to positions spanning from the start to the end of the pullback).
Regarding claim 6, the modified Wang 2015 does not fully teach collecting the plurality of fluorescence data values includes collecting one fluorescence data value for each of the A-lines in 1% of the plurality of B-scan frames of the entire pullback, and wherein calculating the central tendency includes calculating a mean of the discrete or separate fluorescence data values in the 1% of the plurality of B-scan frames. Specifically, the modified Wang 2015 teaches (i) collecting one fluorescence data value per A-line, (ii) grouping A-lines into rotation-synchronized frames, and (iii) acquiring those frames sequentially during an entire pullback (Wang 2015, p. 7, Sec. 2.2.3). However, the modified Wang 2015 does not disclose selecting 1% of the pullback frames and calculating a mean of the fluorescence data values in that 1% subset.
Ishihara teaches that a fluorescence-image processing pipeline may set (or update) a threshold “for every several frames” rather than for every frame, such that a threshold is set only when the frame number reaches an “n-th frame” (Ishihara, ¶[0101]-[0102]). In Ishihara, the threshold-setting event is based on a per-frame mean by teaching that “the threshold value S is set on the basis of the sum of the average gradation value m of the entire image of the Subsequent frame and the standard deviation σ” (Ishihara, ¶[0102]). In Ishihara, the “entire image” refers to the acquired image of the single “Subsequent frame” being processed at that step (not the entire collection of frames), because the threshold update is triggered when the frame number i reaches the “n-th frame” for a Subsequent frame, and then the threshold is set based on that Subsequent frame’s average gradation value m (Ishihara, ¶[0101]-[0102]).
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have further modified the modified Wang 2015 in view of Ishihara to calculate a mean of discrete or separate fluorescence data values using only a subset of pullback frames (i.e., not every frame), by configuring the processor to compute a per-frame mean for periodically selected frames and to use that subset for background/threshold determination (Ishihara, ¶[0101]-[0102]). This modification is feasible because Wang 2015 already digitizes/collects fluorescence values per A-line and organizes them into sequential rotation-synchronized frames during a pullback, and Ishihara’s “every n-th frame” threshold-setting is a software-side selection of a subset of frames combined with a per-frame mean calculation.
Further, selecting a particular fraction such as “1%” corresponds to a routine choice of the update/selection frequency (i.e., setting n so that one frame out of every 100 frames is used), which is an optimization of a result-effective variable balancing computational burden and estimation stability. The benefit of the combination is reducing computation (by calculating means on only a small representative fraction of frames) while still providing a robust frame-based mean suitable for background/threshold setting within the pullback processing pipeline.
Regarding claim 7, Wang 2015 teaches that the one or more frames include one or more B-scan frames acquired by the imaging catheter in a region of low or no fluorescence within the bodily lumen (Wang 2015, Fig. 5, p. 10: “…The color map ranges from blue (low NIRAF intensity) to green, yellow and white (highest NIRAF intensity)”, this shows that some regions of the pullback image correspond to low or no fluorescence, indicating that the underlying B-scan frames also captured regions of low or no fluorescence within the lumen; p. 10, Sec. 3.1: “The x-axis of the 2D NIRAF intensity map corresponds to the longitudinal pullback position, and the y-axis, the scanning angle (i.e., 0 to 360 degrees)”, this explains that the en face map represents a series of B-scan frames acquired sequentially during pullback, confirming that some frames are located in regions of low or no fluorescence; p. 7, Sec. 2.2.3: “Each OFDI frame consisted of 1,024 A-lines, with a frame rate of 39 frames/sec (40 kHz/1024)”, this shows the definition of a frame and supports that multiple such frames are acquired sequentially during pullback, some of which correspond to low- or no-fluorescence regions).
Regarding claim 8, the modified Wang 2015 does not fully teach the fluorescence data values includes collecting one fluorescence data value for each of the A-lines of the one or more B-scan frames, and wherein calculating the central tendency includes calculating an average of the discrete or separate fluorescence data values for each of the one or more B-scan frames, and selecting a lowest average among the averages for the one or more B-scan frames. Specifically, as set forth in claim 1 and claim 7, the modified Wang 2015 teaches collecting one fluorescence data value per A-line and organizing those A-lines into rotation-synchronized B-scan frames acquired sequentially during a pullback (Wang 2015, p. 7, Sec. 2.2.3), and further teaches that such B-scan frames include frames acquired in regions of low or no fluorescence within the bodily lumen, as evidenced by NIRAF intensity maps showing low-intensity regions along the pullback (Wang 2015, Fig. 5; Wang 2015, p. 10, Sec. 3.1). In addition, Ishihara teaches calculating a per-frame central tendency by calculating “an average gradation value m of the entire image of the Subsequent frame” (Ishihara, ¶[0070]), which together provide a framework for (i) identifying frames corresponding to low- or no-fluorescence regions and (ii) calculating a per-frame mean of discrete or separate fluorescence data values for those frames. However, the modified Wang 2015 does not disclose selecting a lowest average among the calculated per-frame averages of the one or more B-scan frames from the low/np fluorescence areas as the background fluorescence threshold.
Ishihara further teaches that the calculation of per-frame averages and the setting of a threshold value are not required to be performed continuously for every acquired frame, but instead may be selectively enabled, disabled, or updated under operator control. Specifically, Ishihara teaches embodiments in which a threshold value is calculated based on a per-frame average and then maintained without recalculation for subsequent frames, as well as embodiments in which the threshold value is updated only when certain conditions are met or when instructed by the operator (Ishihara, ¶[0098]–[0102]). These teachings establish that Ishihara’s processing pipeline supports using a subset of frames to determine a threshold value and then applying that threshold to subsequent frames without further averaging.
In this framework, a clinician may allow per-frame average calculations to occur while imaging a region identified as having low or no fluorescence, as taught by Wang 2015’s NIRAF pullback maps showing regions of low NIRAF intensity along the pullback (Wang 2015, Fig. 5; Wang 2015, p. 10, Sec. 3.1), and then disable further threshold updating once sufficient background frames have been acquired. The per-frame averages calculated during this background-dominant region therefore define a candidate set of average values corresponding to low- or no-fluorescence frames.
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have further modified the modified Wang 2015 in view of Ishihara to select, from among the per-frame averages calculated during such a clinician-selected background region, a lowest average as the background fluorescence threshold. This selection does not alter Ishihara’s principle of operation because Ishihara already (i) uses a per-frame average as the decision statistic for threshold setting and (ii) permits operator-controlled threshold updating and holding; selecting the lowest average among values already calculated within Ishihara’s framework is merely a conservative selection criterion applied to those existing values.
Ughi 2016 further confirms, in the same OCT–NIRAF field, the routine practice of using the minimum signal derived from an acquired dataset as a background reference for normalization (Ughi 2016, p. 4, DATA PROCESSING), reinforcing that selecting the lowest per-frame average from background frames as the background fluorescence threshold is consistent with established fluorescence imaging practices.
The benefit of the combination is improved robustness in subsequent fluorescence image thresholding and contrast enhancement by allowing a clinician to derive a conservative background fluorescence threshold from frames acquired in low- or no-fluorescence regions and to apply that threshold consistently during subsequent imaging, while retaining the background removal processing taught by the modified Wang 2015.
Regarding claim 9, the modified Wang 2015 does not fully teach that the one or more frames includes a first B-scan frame acquired by the imaging catheter from a first region within the bodily lumen, wherein collecting the plurality of fluorescence data values includes collecting one fluorescence data value for each of the A-lines of the first B-scan frame acquired from the first region, and wherein calculating the central tendency includes calculating an average of the discrete or separate fluorescence data values for the first B-scan frame. As shown in claim 1, the modified Wang 2015 already teaches catheter pullback imaging that yields rotation-synchronized frames along the bodily lumen and fluorescence data values aligned to individual A-lines within those frames (Wang 2015, p. 7, Sec. 2.2.3; Wang 2015, p. 8, Sec. 2.2.5). It further teaches that frames acquired during pullback correspond to different longitudinal positions within the lumen, i.e., different regions, as reflected by the pullback axis of the two-dimensional NIRAF intensity map (Wang 2015, p. 10: “The x-axis of the 2D NIRAF intensity map corresponds to the longitudinal pullback position”), and illustrates using a specific pullback location as a region of interest associated with a specific OCT-NIRAF frame (Wang 2015, p. 10: “taken at the location of the dashed line in Fig. 5”). However, while the modified Wang 2015 establishes that frames inherently correspond to different regions within the bodily lumen, it does not teach, as an explicit processing choice, selecting a particular region-designated frame, such as a first region, and applying the per-frame central tendency calculation specifically to the B-scan frame acquired from that first region.
Ishihara teaches that frame-level threshold calculations are not required to be applied uniformly to all frames, but may instead be selectively applied to particular frames as observation proceeds. In particular, Ishihara teaches configuring threshold setting to occur “for every several frames” and resetting the threshold when a frame number reaches an “n-th frame” (Ishihara, ¶[0101]: “a threshold value may be set for every several frames”, “the threshold value should be changed once when a frame number i reaches an n-th frame”; ¶[0102]: “when the frame number i of a Subsequent frame … reaches the n-th frame … steps SB2 to SD7 may be repeated”). Ishihara further teaches that the statistic used for such threshold setting is an average computed for the “entire image of the Subsequent frame,” meaning the entire image of an individual frame rather than an entire multi-frame pullback dataset (Ishihara, ¶[0070]: “calculates an average gradation value m of the entire image of the Subsequent frame”). Together, these teachings demonstrate that it is appropriate to designate particular frames, including an initial or first set of frames corresponding to an initial region during observation, as the frames to which the average-based central tendency calculation is applied. Ishihara therefore supplies the missing teaching that frame-level averaging may be selectively applied to frames associated with a designated region, consistent with selecting a first region in Wang’s pullback and applying the average calculation to the B-scan frame acquired from that region.
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have further modified the modified Wang 2015 in view of Ishihara to select a first region within the bodily lumen during pullback and to calculate the central tendency as an average for the rotation-synchronized B-scan frame acquired from that first region. This modification is feasible because the modified Wang 2015 already associates frames with pullback position and collects discrete A-line-aligned fluorescence values within each frame, and Ishihara teaches frame-indexed processing in which average-based calculations are performed only at selected frames, such as at initial frames or at frames designated by progression of observation. The benefit is improved robustness of threshold determination by reducing the influence of localized outliers and pullback-to-pullback variability through region-specific frame selection for central tendency calculation, thereby improving consistency of fluorescence quantification across different lumen regions.
Regarding claim 10, the modified Wang 2015 does not fully teach that the one or more frames further include a second B-scan frame acquired by the imaging catheter from a second region different from the first region within the bodily lumen, wherein collecting the plurality of fluorescence data values further includes collecting one fluorescence data value for each of the A-lines of the second B-scan frame acquired from the second region, wherein calculating the central tendency further includes calculating an average of the discrete or separate fluorescence data values for the second B-scan frame, and wherein assigning the central tendency of one of the one or more frames as the background fluorescence threshold includes assigning a lowest central tendency among the central tendencies of the first B-scan frame and the second B-scan frame as the background fluorescence threshold. As shown in claim 1, the modified Wang 2015 already teaches catheter pullback imaging that yields rotation-synchronized frames along the bodily lumen and fluorescence data values aligned to individual A-lines within those frames (Wang 2015, p. 7, Sec. 2.2.3; Wang 2015, p. 8, Sec. 2.2.5). As further shown in claim 9, the modified Wang 2015 teaches that frames correspond to longitudinal pullback position, which is a spatial region within the lumen (Wang 2015, p. 10: “The x-axis of the 2D NIRAF intensity map corresponds to the longitudinal pullback position”), and illustrates that different pullback locations correspond to different frames associated with different regions (Wang 2015, p. 10: “taken at the location of the dashed line in Fig. 5”; “taken at the location of the dotted line in Fig. 5”). However, the modified Wang 2015 does not teach the specific decision logic of comparing central tendencies calculated for frames acquired at different longitudinal pullback positions and assigning the lowest of those frame-level central tendencies as the background fluorescence threshold.
Ishihara teaches performing a frame-level average calculation for an individual frame image (Ishihara, ¶[0070]: “calculates an average gradation value m of the entire image of the Subsequent frame”), and further teaches that threshold setting may be performed selectively at particular frames rather than indiscriminately for every frame, including resetting the threshold at defined frame intervals as observation proceeds (Ishihara, ¶[0101]: “a threshold value may be set for every several frames”; ¶[0101]: “the threshold value should be changed once when a frame number i reaches an n-th frame”; ¶[0102]: “when the frame number i of a Subsequent frame … reaches the n-th frame … steps SB2 to SD7 may be repeated”). Ishihara is relied upon here solely for the teachings of calculating a frame-level central tendency for an individual frame and selectively determining which frames are used to update threshold calculations, and does not define or rely on spatial regions within the lumen.
Ughi 2016 teaches that quantitative NIRAF data may be normalized using minimum values acquired across imaging data, demonstrating the use of a lowest-value reference as a conservative baseline for fluorescence signal evaluation (Ughi 2016, p. 1306, DATA PROCESSING: “Quantitative NIRAF data were normalized between 0 and 1 using the minimum and maximum NIRAF values acquired in the study”). Accordingly, once per-frame averages are calculated for frames acquired at different longitudinal pullback positions in the modified Wang 2015 using the frame-level averaging taught by Ishihara, it would have been an expected and routine selection operation to compare the resulting computed frame-level central tendencies and assign the lowest value as a conservative background fluorescence threshold, consistent with Ughi’s use of minimum-based normalization, and without altering Ishihara’s frame-level processing because Ishihara is relied upon only to compute the per-frame average and to support performing that computation at selected frames.
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the modified Wang 2015 in view of Ishihara and Ughi 2016 to assign a lowest central tendency among the central tendencies of the first B-scan frame and the second B-scan frame as the background fluorescence threshold. This modification is feasible because the modified Wang 2015 already produces sequential, region-correlated frames with per-A-line fluorescence values, Ishihara teaches calculating frame-level averages and selectively updating threshold calculations at designated frames during observation, and Ughi 2016 demonstrates using minimum values from acquired NIRAF data as a conservative low reference. The benefit is improved reliability and stability of background fluorescence estimation across spatially distinct lumen regions by reducing sensitivity to localized fluorescence elevations, plaque-related signal variation, and transient acquisition artifacts, thereby improving consistency of fluorescence quantification as the catheter progresses through different regions of the bodily lumen.
Regarding claim 11, Wang 2015 teaches that a catheter-based imaging system for quantifying fluorescence emitted from within a bodily lumen (Wang 2015, Abstract: “In this study, we used a recently developed imaging system and double-clad fiber (DCF) catheter capable of simultaneously acquiring both OCT and red excited near-infrared autofluorescence (NIRAF) images (excitation: 633 nm, emission: 680nm to 900nm)... We found that NIRAF is elevated in lesions that contain necrotic core – a feature that is critical for vulnerable plaque diagnosis... These results suggest that multimodality intracoronary OCT-NIRAF imaging technology may be used in the future to provide improved characterization of coronary artery disease in human patients”, this shows Wang 2015 teaches a catheter-based imaging system for quantifying fluorescence emitted from within a bodily lumen); an imaging catheter that scans the bodily lumen... with light of a wavelength capable of stimulating emission of fluorescence from within the bodily lumen (Wang 2015, p. 4, Sec. 2.1.1: “The multimode inner cladding (NA ≥ 0.46, diameter = 124–126 µm, fiber outer diameter = 250 µm) was used to guide 633 nm excitation light and to collect NIRAF tissue emission...”, which teaches that the catheter transmits excitation light at 633 nm to stimulate fluorescence emission from within tissue regions of the bodily lumen); and a processor (Wang 2015, p. 7, Sec. 2.2.3: “The sampling rate of NIRAF signal was 40 kHz, enabling integration of the NIRAF emission within the period of a single A-line. Each OFDI frame consisted of 1,024 A-lines, with a frame rate of 39 frames/sec (40 kHz/1024). The rotary junction was operated at 39 rotations per second so each OFDI image covered one rotation”; p. 7–8, Sec. 2.2.5: “The OCT background images of PBS solution were averaged and subtracted from each OCT image... the NIRAF background signal... was also averaged and subtracted”, Wang 2015’s disclosure of frame construction, averaging, subtraction, normalization, and statistical analysis of fluorescence data values necessarily requires a processor executing stored instructions, and therefore teaches a processor configured to perform the claimed functions under the broadest reasonable interpretation) configured to: collect a plurality of fluorescence data values corresponding to the fluorescence emitted from different regions within the bodily lumen while the scanning is being performed... (Wang 2015, p. 7, Sec. 2.2.3: “The sampling rate of NIRAF signal was 40 kHz, enabling integration of the NIRAF emission within the period of a single A-line. Each OFDI frame consisted of 1,024 A-lines, with a frame rate of 39 frames/sec (40 kHz/1024). The rotary junction was operated at 39 rotations per second so each OFDI image covered one rotation”, which teaches that fluorescence data values are acquired in real-time during scanning at the A-line level and organized into frames corresponding to catheter rotation, thereby teaching a plurality of fluorescence data values corresponding to emission from different regions of the bodily lumen during the scanning process, the acquisition of which is performed by a processor executing instructions under the broadest reasonable interpretation); group the plurality of collected fluorescence data values into one or more frames, each frame of the one or more frames having a number of discrete fluorescence data values of the plurality of fluorescence data values collected from one full rotation of the imaging catheter within the bodily lumen (Wang 2015, p. 7, Sec. 2.2.3: “The sampling rate of NIRAF signal was 40 kHz, enabling integration of the NIRAF emission within the period of a single A-line. Each OFDI frame consisted of 1,024 A-lines, with a frame rate of 39 frames/sec (40 kHz/1024). The rotary junction was operated at 39 rotations per second so each OFDI image covered one rotation”, which teaches that each frame comprises individual A-line-based fluorescence measurements that are separately acquired within a single rotation, such that the fluorescence data values within a frame are discrete from one another and from fluorescence data values of other frames, the grouping of which is performed by a processor executing instructions under the broadest reasonable interpretation); and the discrete fluorescence data values of one frame of the one or more frames being discrete from fluorescence data values of each of the other frames of the one or more frames (Wang 2015, p. 7, Sec. 2.2.3: “Each OFDI frame consisted of 1,024 A-lines, with a frame rate of 39 frames/sec (40 kHz/1024). The rotary junction was operated at 39 rotations per second so each OFDI image covered one rotation”, this teaches that each frame is a distinct set of A-line based data acquired during a single rotation, such that the fluorescence data values of one frame are separate from the fluorescence data values of other frames corresponding to other rotations, as managed by a processor executing instructions under the broadest reasonable interpretation).
Also regarding claim 11, Wang 2015 does not fully teach performing the functions of scanning, collecting, and calculating while the scanning is being performed in the in vivo environment with an imaging catheter. Rather, Wang 2015 teaches imaging and fluorescence data acquisition using a catheter system, but the imaging is performed on ex vivo human cadaver coronary arteries (Wang 2015, p. 8, Sec. 2.2.6: “A total of 15 coronary arteries from 5 human cadaver hearts were imaged using the multimodality OCT-NIRAF system and catheter”). Wang 2015 further indicates that the disclosed imaging technique is intended for clinical application, stating that “These results suggest that multimodality intracoronary OCT-NIRAF imaging technology may be used in the future to provide improved characterization of coronary artery disease in human patients” (Wang 2015, Abstract), thereby indicating that the ex vivo imaging is performed as validation of a system intended for in vivo use. However, it does not explicitly teach scanning, collecting, and calculating in an in vivo environment.
Ughi 2016 teaches that “The authors present the clinical imaging of human coronary arteries in vivo using a multimodality optical coherence tomography (OCT) and near-infrared autofluorescence (NIRAF) intravascular imaging system and catheter” (Ughi 2016, Abstract) and further teaches that “OCT-NIRAF imaging of the vessel undergoing PCI was performed in all cases” (Ughi 2016, p. 1306), thereby demonstrating that the imaging catheter itself is operated in an in vivo environment to perform fluorescence imaging during active pullback scanning in human patients, and thus evidencing that both data acquisition (scanning) and associated processing occur while the catheter is actively scanning the bodily lumen in vivo.
Ishihara teaches that fluorescence image processing is performed on a per-frame basis as successive frames are generated, including calculating an “average gradation value m (=1050) of the entire image” (Ishihara, ¶[0083]) and an “average gradation value m (=1365) of the entire image of the subsequent frame” (Ishihara, ¶[0088]), and further teaches that processed images are immediately displayed such that “a new corrected fluorescence image... is displayed on the monitor 50” (Ishihara, ¶[0084]). Because each frame corresponds to image data generated during ongoing acquisition, and Ishihara performs calculations and display updates upon generation of each successive frame, the processing is performed contemporaneously with acquisition of the frames, which in the context of catheter pullback as evidenced by Ughi 2016 corresponds to processing during scanning. Ishihara additionally teaches that an operator may adjust a threshold value during generation of corrected fluorescence images of subsequent frames, thereby evidencing that processing is performed during ongoing frame generation.
It would have been prima facie obvious before the effective filing date of the claimed invention to modify Wang 2015 in view of Ughi 2016 and Ishihara to perform the scanning of the bodily lumen in an in vivo environment and to perform the processing steps (collecting and calculating) during the scanning process because Ughi 2016 demonstrates that the same type of catheter-based OCT-NIRAF imaging system is used clinically in vivo, and Ishihara demonstrates that fluorescence image data is processed on a per-frame basis as frames are generated and displayed, including operator interaction during ongoing image generation, thereby evidencing real-time or near real-time processing of acquired image data. Adapting Wang’s catheter-based imaging technique for in vivo use and applying Ishihara’s per-frame processing during acquisition represents a predictable use of known imaging and data processing techniques to enable real-time or near real-time evaluation of fluorescence data during catheter-based imaging while maintaining the underlying catheter-based OCT-NIRAF imaging architecture of Wang 2015, including its excitation, detection, and frame-based acquisition processes. Accordingly, the modification does not alter the fundamental principle of operation of Wang 2015 but instead applies known processing techniques within its existing framework. The benefit of the combination is that performing fluorescence data processing during in vivo scanning enables real-time or near real-time evaluation of fluorescence signals, thereby improving clinical assessment of tissue characteristics during the procedure, with the processing performed by a processor executing instructions under the broadest reasonable interpretation.
Also regarding claim 11, the modified Wang 2015 does not fully teach calculate a central tendency of the discrete fluorescence data values in each of the one or more frames while the scanning is being performed in the in vivo environment. Rather, the modified Wang 2015 teaches collecting fluorescence intensity data in rotation-synchronized frames and performing normalization of fluorescence data, but it does not teach calculating a central tendency, such as an average or mean, of the discrete fluorescence data values in each frame while the scanning is being performed in the in vivo environment.
Ishihara teaches calculating an “average gradation value m (=1050) of the entire image” (Ishihara, ¶[0083]) and further calculating an “average gradation value m (=1365) of the entire image of the subsequent frame” (Ishihara, ¶[0088]), thereby teaching that Ishihara calculates a central tendency for each individual frame as it is acquired — the initial frame (¶[0083]) and each subsequent frame (¶[0088]) — directly addressing the “each of the one or more frames” limitation, which is performed by a processor executing instructions under the broadest reasonable interpretation. Because the gradation values of Ishihara correspond to pixel intensity values representing fluorescence emission at each pixel, and each image corresponds to a frame of such values, the calculated average directly corresponds to a central tendency of discrete fluorescence data values within a frame under the broadest reasonable interpretation.
Ughi 2016 teaches that sequential OCT-NIRAF frames are acquired during active in vivo catheter pullback (Ughi 2016, Abstract; p. 1306), thereby evidencing that frame generation occurs while the imaging catheter is scanning the bodily lumen in an in vivo environment. Accordingly, applying Ishihara’s per-frame central tendency calculation to such sequentially acquired frames necessarily results in the central tendency being calculated while the scanning is being performed in the in vivo environment.
It would have been prima facie obvious before the effective filing date of the claimed invention to have further modified the modified Wang 2015 in view of Ishihara to calculate a central tendency of the discrete fluorescence data values in each of the one or more frames while the scanning is being performed in the in vivo environment because frame-based statistical summarization of image intensity data is a well-known and predictable processing step used to reduce data dimensionality and enable subsequent frame-level comparisons and thresholding operations, and Ishihara explicitly demonstrates such frame-level statistical processing applied to sequential image frames. This modification would have been possible because Wang 2015 already forms rotation-based frames from discrete fluorescence measurements, Ishihara expressly performs a per-frame average calculation upon acquisition of each subsequent frame, and Ughi 2016 evidences that such catheter-based fluorescence imaging is performed in vivo during pullback scanning. The benefit of the combination is that a frame-level central tendency provides a consistent statistical basis for subsequent frame-level threshold and background processing while remaining within the existing catheter imaging pipeline. Further, because Ishihara performs the average calculation upon acquisition of each subsequent frame and Ughi 2016 demonstrates that such frames are acquired during active catheter pullback, the calculation of the central tendency necessarily occurs while the scanning is being performed in the in vivo environment.
Also regarding claim 11, the modified Wang 2015 does not fully teach assign the central tendency of one of the one or more frames as a background fluorescence threshold for the in vivo environment by determining a lowest central tendency of the one or more frames via, until the scanning is finished, setting a first central tendency of a first frame of the one or more frames as a pending central tendency, iteratively comparing the respective central tendency of each of the following one or more frames to the pending central tendency, setting any respective central tendency that is lower than the pending central tendency as the pending central tendency, and once the scanning has finished assigning the pending central tendency as the lowest central tendency and as the background fluorescence threshold for the in vivo environment. Rather, the modified Wang 2015 teaches collecting fluorescence data in rotation-synchronized frames during catheter pullback, performing per-frame central tendency calculations as modified above in view of Ishihara, and using fluorescence intensity values as part of normalization processing in fluorescence imaging, but the modified Wang 2015 does not fully teach the specifically recited iterative pending-minimum process in which a first frame central tendency is set as a pending central tendency, each subsequent frame central tendency is iteratively compared to the pending central tendency during scanning, any lower central tendency replaces the pending central tendency, and the pending central tendency is assigned at the end of scanning as the lowest central tendency and background fluorescence threshold.
Ishihara teaches calculating per-frame average values for sequentially acquired frames (Ishihara, ¶[0083], ¶[0088]).
Ughi 2016 evidences that fluorescence intensity varies across frames during in vivo acquisition, including lower-intensity measurements (Ughi 2016, p. 1306). The gradation values processed by Ishihara correspond to pixel intensity values of an image, which are directly analogous to fluorescence intensity values in Wang 2015, as both represent scalar intensity measurements associated with image data. Specifically, Ishihara’s corrected fluorescence image is generated by dividing the fluorescence image by a white-light reference image (Ishihara, ¶[0046]), such that the gradation values in the corrected fluorescence image are quantitative representations of fluorescence emission intensity at each pixel, normalized for observation distance and angle. These gradation values represent the same physical quantity as the fluorescence data values of Wang 2015, expressed in digital units, and statistical processing applied to such values is equivalently applicable across both systems.
Ma teaches that “A background correction method for a fluorescence microscopy system includes receiving a raw image stack, determining a number of temporal minimum intensity values for each pixel location from the raw image stack, and calculating an expected background value for each pixel location based on the number of temporal minimum intensity values for the pixel location” (Ma, Abstract). Ma is in the analogous field of fluorescence image processing and background correction, which involves the same problem of distinguishing signal from background intensity in fluorescence imaging data as in Wang 2015. Ma further teaches “RECEIVE RAW IMAGE STACK", “SEGMENT THE IMAGE STACK INTO A NUMBER OF SUB-STACKS", “CALCULATE THE MINIMUM VALUE OF THE PIXELS ALONG THE TEMPORAL AXIS FOR EACH PIXEL LOCATION FOR EACH SUB-STACK", and “CALCULATE THE EXPECTED BACKGROUND VALUE FOR EACH PIXEL LOCATION ACCORDING TO A DERIVED TRANSFER FUNCTION BASED ON THE OBTAINED MINIMUM VALUES FOR EACH PIXEL LOCATION” (Ma, Fig. 3), thereby teaching determining minimum intensity values over time and using those values to calculate background intensity A person of ordinary skill in the art would have recognized that Ma’s use of temporal minimum intensity values to estimate background conditions at the pixel level applies equivalently to frame-level statistical representations in a catheter pullback context, where each frame corresponds to a different spatial region and per-pixel temporal tracking is not feasible. It would have been obvious to adapt Ma’s minimum-based background determination to operate on per-frame central tendencies calculated by Ishihara by identifying the lowest per-frame central tendency across the pullback as the background fluorescence reference, which represents a predictable application of Ma’s minimum-based background estimation principle using a more statistically robust frame-level metric. The initialization step of setting the first central tendency as the pending central tendency is supported by Ishihara’s FIG. 7 process, in which the threshold-value setting unit acquires the first corrected fluorescence image, calculates the average gradation value m (=1050) of that initial frame, and stores it as the operative baseline value from which subsequent comparisons proceed (Ishihara, ¶[0083]).
Lemire teaches that computing a minimum over a sequence is performed via comparisons across the values of the sequence, stating that “Computing either the global maximum or the global minimum of an array of n elements requires n − 1 comparisons” (Lemire, p. 1), and further teaches that such running minimum computations can be performed with zero stream latency, defined as “the maximum number of data points required after the window has passed” (Lemire, Table I), thereby evidencing that minimum determination may be implemented through sequential comparison of incoming data values as they are acquired without requiring future data. This inherently corresponds to an iterative process in which an initial value is selected, each subsequent value is compared to the current minimum, the current minimum is updated when a lower value is identified, and the final minimum is obtained after processing all values in the sequence.
It would have been prima facie obvious before the effective filing date of the claimed invention to have further modified the modified Wang 2015 in view of Ishihara, Ughi 2016, Ma, and Lemire to assign the central tendency of one of the one or more frames as a background fluorescence threshold for the in vivo environment by determining a lowest central tendency using the claimed iterative comparison process. This modification would have been possible because the modified Wang 2015 already provides rotation-synchronized frame data acquired during catheter pullback, Ishihara already calculates a per-frame central tendency for each acquired frame, Ma teaches determining minimum intensity values over time and using those values to calculate background intensity (Ma, Abstract; Fig. 3), and Lemire teaches sequential comparison of values to determine a minimum over a sequence. Further, assigning the lowest value after scanning the completes is implicit because the determination of a minimum across a sequence of values requires processing the entire sequence, and the final minimum cannot be known until all values have been evaluated, wherein the catheter pullback defines a finite sequence of frames that must be fully processed before the lowest central tendency can be determined. The benefit of the combination is improved robustness in determining a background fluorescence threshold from in vivo sequential frame data by using an iterative lowest-value tracking process that identifies the lowest background-dominant frame statistic across the scan. The combination of Ishihara’s per-frame central tendency calculation, Ma’s use of minimum values across temporally acquired image data for background estimation, and Lemire’s sequential comparison methodology collectively yields the iterative process of initializing a pending value, comparing each subsequently computed frame-level value, updating the pending value when a lower value is identified, and finalizing the lowest value after completion of data acquisition, which represents a predictable integration of known data processing techniques performed by a processor executing instructions under the broadest reasonable interpretation.
Also regarding claim 11, the modified Wang 2015 does not fully teach adjust the fluorescence data values of the one or more frames based on the background fluorescence threshold for the in vivo environment by subtracting a background fluorescence value corresponding to the background fluorescence threshold from each of the fluorescence data values or by removing any fluorescence data value from the fluorescence data values that is equal to or lower than the background fluorescence threshold, thereby generating adjusted fluorescence data values. Rather, the modified Wang 2015 teaches adjusting fluorescence data by subtracting a PBS-derived background signal and, as modified above, teaches determining a background fluorescence threshold from frame-based processing, but the modified Wang 2015 does not fully teach the specifically recited alternatives of adjusting the fluorescence data values based on the determined background fluorescence threshold by subtracting a corresponding background fluorescence value from each fluorescence data value or by removing any fluorescence data value from the fluorescence data values that is equal to or lower than the background fluorescence threshold.
Wang teaches subtracting background fluorescence using PBS-derived background values (p. 7–8, Sec. 2.2.5). Ishihara teaches threshold-based adjustment including replacing values below a threshold with zero (¶[0084]: “the image adjuster 51 replaces the gradation values of pixels having gradation values smaller than the threshold value S (=1441) with zero”; ¶[0100]: “the image adjuster 51 may adjust the gradation values of the corrected fluorescence image on the basis of the current threshold value S”).
It would have been prima facie obvious before the effective filing date of the claimed invention to have further modified the modified Wang 2015 in view of Ishihara to adjust the fluorescence data values of the one or more frames based on the background fluorescence threshold for the in vivo environment by subtracting a background fluorescence value corresponding to the background fluorescence threshold from each fluorescence data value or by removing any fluorescence data value from the fluorescence data values that is equal to or lower than the background fluorescence threshold, thereby generating adjusted fluorescence data values. This modification would have been possible because Wang already performs background subtraction from fluorescence data and Ishihara teaches threshold-based removal of lower-valued fluorescence image data. The benefit of the combination is improved suppression of background-level fluorescence contributions while retaining lesion-related fluorescence signal, thereby improving contrast and interpretability in the resulting fluorescence images, as implemented by a processor executing instructions under the broadest reasonable interpretation.
Also regarding claim 11, the modified Wang 2015 does not fully teach generate an image based on the adjusted fluorescence data values. Rather, the modified Wang 2015 teaches generating fluorescence images from processed fluorescence data and, as modified above, teaches adjusting fluorescence data values based on a background fluorescence threshold, but the modified Wang 2015 does not fully teach generating an image specifically based on fluorescence data values adjusted using the claimed threshold-based processing.
Wang 2015 teaches that “The normalized NIRAF signal was rendered as a ring shaped image, where each NIRAF data point matches the corresponding OCT A-line at each given rotational position of the catheter” (Wang 2015, p. 7–8).
Ishihara teaches generating and displaying corrected fluorescence images after threshold-based adjustment (¶[0084]: “a new corrected fluorescence image, in which 91.8% of the display of the background is eliminated and 89.5% of the display of the lesion is maintained, is displayed on the monitor 50”; ¶[0100]: “the image adjuster 51 may adjust the gradation values of the corrected fluorescence image on the basis of the current threshold value S”).
It would have been prima facie obvious before the effective filing date of the claimed invention to have further modified the modified Wang 2015 in view of Ishihara to generate an image based on the adjusted fluorescence data values because Ishihara explicitly uses threshold-adjusted fluorescence image values to generate and display a corrected fluorescence image. The benefit of the combination is improved visual contrast and suppression of background fluorescence in the generated image while preserving lesion-related fluorescence signal.
Regarding claim 12, the modified Wang 2015 teaches that the fluorescence data values correspond to one or more of an intensity, an amplitude, and a lifetime of the fluorescence (Wang 2015, p.3, Sec. 2: “Scanning the tissue en face generated a NIRAF intensity map”, p. 7–8, Sec. 2.2.5–2.2.6: “The normalized NIRAF signal was rendered as a ring shaped image, where each NIRAF data point matches the corresponding OCT A-line at each given rotational position of the catheter”, this shows that NIRAF signal values are the measured fluorescence intensity used for image generation; p. 4-5, Sec. 2.1.2: “The collected NIRAF light (emission wavelength range 675nm to 900nm) was focused onto a photomultiplier tube (PMT, model H-5784, Hamamatsu, Japan) … The PMT output was digitized by a data acquisition board (PCA-6110, National Instruments, Texas, USA)”, this shows acquisition of fluorescence signal values for quantitation, i.e., intensity).
Regarding claim 13, the modified Wang 2015 teaches that the fluorescence data values include N detected fluorescence data values, where N is a positive integer greater than 2 (Wang 2015, p. 7, Sec. 2.2.3: “The sampling rate of NIRAF signal was 40 kHz, enabling integration of the NIRAF emission within the period of a single A-line. Each OFDI frame consisted of 1,024 A-lines, with a frame rate of 39 frames/sec (40 kHz/1024). The rotary junction was operated at 39 rotations per second so each OFDI image covered one rotation”, this teaches that for each rotation-synchronized frame the system detects a plurality of discrete fluorescence data values corresponding to individual A-lines, such that N is a positive integer greater than 2).
Also regarding claim 13, the modified Wang 2015 does not fully teach that the processor calculates one or more of an average, a mean, or a median as the central tendency of the discrete or separate fluorescence data values. The modified Wang 2015, as established in claim 11, teaches detecting and processing a large number of discrete fluorescence data values per rotation-synchronized frame (1,024 A-lines per frame), grouping those values into frames corresponding to full catheter rotations, and operating on those frame-level fluorescence datasets for background handling and image generation, thereby satisfying the requirement that the fluorescence data values include N detected values where N is a positive integer greater than 2 (Wang 2015, p. 7, Sec. 2.2.3; p. 7–8, Sec. 2.2.5). It was additionally shown to teach using an average (mean) for a frame as a central tendency by applying Ishihara’s per-frame “average gradation value m of the entire image of the Subsequent frame” to Wang’s rotation-synchronized frames (Ishihara, ¶[0070]; Wang 2015, p. 7, Sec. 2.2.3).
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have further modified the modified Wang 2015 in view of Ishihara to cause the processor to calculate one or more of an average, a mean, or a median as the central tendency of the detected fluorescence data values for a rotation-synchronized frame. This modification is feasible because the modified Wang 2015 already forms frames from many discrete fluorescence measurements, and Ishihara teaches a per-frame average computation applied to the entire image of a frame using routine image-processing operations. The predictable benefit of this modification is a statistically representative frame-level value that improves robustness and consistency in subsequent fluorescence processing and threshold-related operations without altering the catheter hardware or acquisition procedure.
Regarding claim 14, the modified Wang 2015 partially teaches that the processor is further configured to: discard the fluorescence data values that are equal to or lower than the background fluorescence threshold; and generate the image based on the fluorescence data values that are larger than a value of the background fluorescence threshold. Specifically, the modified Wang 2015 teaches collecting and processing fluorescence data values for catheter-based imaging, including performing background removal by subtracting a background signal and rendering the processed NIRAF signal as an image that is co-registered to OCT A-lines, as established in claim 11 (Wang 2015, p. 7–8, Sec. 2.2.5; Wang 2015, p. 8–9, Sec. 2.2.5–2.2.6). However, the modified Wang 2015 does not specify discarding fluorescence data values that are equal to or lower than a background fluorescence threshold and generating the image based on only those fluorescence data values that are larger than the background fluorescence threshold.
Ishihara teaches generating a corrected fluorescence image by suppressing values at or below a threshold value S, specifically teaching that “the image adjuster 51 replaces the gradation values of pixels having gradation values Smaller than the threshold value S (=1575) with zero”, which discards or suppresses subthreshold fluorescence-related values so the displayed corrected fluorescence image is generated based on values above the threshold (Ishihara, ¶[0065]). Ishihara further teaches that the corrected fluorescence image is displayed with increased contrast after this threshold-based suppression (Ishihara, ¶[0066]: “a new corrected fluorescence image with increased contrast between the area displaying the lesion and the area displaying the background is displayed on the monitor 50”).
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the modified Wang 2015 in view of Ishihara to configure the processor to discard fluorescence data values that are equal to or lower than the background fluorescence threshold and to generate the image based on the fluorescence data values that are larger than the background fluorescence threshold by applying a background fluorescence threshold to suppress fluorescence data values at or below the threshold (e.g., setting values at or below the threshold to zero) and generating the image from the remaining values above the threshold. This modification is feasible because the modified Wang 2015 already renders fluorescence images from per-rotation frame data after post-processing, and Ishihara’s threshold-based suppression is a software-side masking operation applied to already-generated fluorescence values without changing catheter structure, rotation rate, pullback acquisition, or image rendering architecture. The benefit of the combination is improved image contrast and robustness by suppressing background-level fluorescence contributions and emphasizing fluorescence signal associated with lesions during display.
Regarding claim 15, the modified Wang 2015 does not fully teach that the processor assigns the central tendency of a portion of the discrete fluorescence data values as the background fluorescence threshold of all of the fluorescence data values where the portion of the discrete fluorescence data values is: (i) less than all of the discrete fluorescence data values, or (ii) less than all of the discrete fluorescence data values and includes a set or predetermined number of lowest fluorescence data values. Rather, the modified Wang 2015 teaches acquiring fluorescence data values that are matched to OCT A-lines within rotation-synchronized frames, deriving a background fluorescence threshold from acquired fluorescence image data, and applying that threshold at a frame level during image adjustment and rendering, as established in claim 11. However, the modified Wang 2015 does not explicitly teach assigning the central tendency of only a portion of the discrete fluorescence data values, where the portion is less than all of the data values or specifically corresponds to a subset such as a predetermined number of lowest fluorescence data values, and further applying that resulting threshold across all fluorescence data values in the dataset. With respect to alternative (i), while Wang 2015 alone does not explicitly teach the portion concept, the combination as applied to claim 11 inherently satisfies this limitation as discussed below.
With respect to alternative (i), the combination as applied to claim 11 already teaches assigning the lowest per-frame central tendency as the background fluorescence threshold for all fluorescence data values, where that central tendency is derived from the discrete fluorescence data values of one frame. Because the fluorescence data values of a single frame represent only a portion of all fluorescence data values collected during a full pullback scan, this teaching implicitly satisfies assigning a central tendency derived from a portion less than all of the discrete fluorescence data values as the background fluorescence threshold of all fluorescence data values.
Ma teaches that background estimation may be based on minimum intensity values over time, including determining minimum or lowest-valued intensity data points to represent background conditions (Ma, Abstract; Fig. 3). Ma further teaches segmenting image data into sub-stacks comprising a defined number of frames prior to performing minimum-based background estimation (Ma, Fig. 3), thereby directly teaching selection of a predetermined number of data points as the basis for background estimation. These teachings establish that selecting a subset of lowest-valued data, including a predetermined number of lowest values, provides a robust and noise-resistant estimation of background conditions in fluorescence imaging data. Accordingly, Ma supports selecting a set or predetermined number of lowest fluorescence data values for use in determining a background-related metric.
It would have been prima facie obvious before the effective filing date of the claimed invention to have further modified the modified Wang 2015 in view of Ma to assign the central tendency of a portion of the discrete fluorescence data values as the background fluorescence threshold of all of the fluorescence data values by deriving a threshold from a selected subset of fluorescence data values. With respect to alternative (i), this modification is already taught by the combination applied to claim 11, which assigns a lowest per-frame central tendency derived from a portion of the data. With respect to alternative (ii), Ma teaches selecting lowest-valued intensity data points and further teaches performing such selection over defined subsets of data (e.g., sub-stacks of predetermined size), and a person of ordinary skill in the art would have recognized that selecting a predetermined number of such lowest values represents a predictable variation of Ma’s disclosed technique to improve robustness and stability of background estimation. This modification would have been possible because the modified Wang 2015 already incorporates frame-based fluorescence data acquisition and threshold determination, and Ma provides a known technique for selecting lowest-valued data for background estimation. The benefit of the combination is improved robustness and stability of background fluorescence threshold determination by basing the threshold on representative low-intensity data while ensuring consistent application across the entire dataset, thereby reducing sensitivity to noise and local signal variation.
Regarding claim 16, Wang 2015 teaches that the processor (i) receives a first N detected discrete fluorescence values (Wang 2015, p. 7, Sec. 2.2.3: “The sampling rate of NIRAF signal was 40 kHz, enabling integration of the NIRAF emission within the period of a single A-line. Each OFDI frame consisted of 1,024 A-lines, with a frame rate of 39 frames/sec (40 kHz/1024). The rotary junction was operated at 39 rotations per second so each OFDI image covered one rotation.”; Wang 2015, p. 6, Sec. 2.2.2: “Signals from the dual balanced detector and the PMT were simultaneously digitized for subsequent processing.”, which teach that fluorescence values are sequentially detected, digitized, and received for processing, thereby teaching receipt of a first N detected discrete fluorescence values by a processor under the broadest reasonable interpretation); but does not fully teach (ii) calculates a mean of the first N detected discrete fluorescence values where the mean is set as the first central tendency. Rather, as established in claim 11, the modified Wang 2015 teaches receiving fluorescence data values sequentially during catheter-based acquisition and organizing those values into frame-based datasets suitable for statistical processing, including calculating central tendency values such as averages for frames of acquired data. However, the modified Wang 2015 does not explicitly teach calculating a mean from an initial subset of detected fluorescence values and setting that mean as an initial or first central tendency used as a reference value.
Ishihara teaches that when a corrected fluorescence image of a frame is acquired, “the threshold-value setting unit 45 calculates an average gradation value m of the entire image” (Ishihara, ¶[0083]) and further teaches calculating such average values for subsequent frames (Ishihara, ¶[0088]). Ishihara additionally teaches storing and comparing calculated average values across frames (Ishihara, ¶[0094]), thereby establishing a processing paradigm in which a calculated mean value is stored and used as a reference central tendency against which subsequently calculated values are compared.
It would have been prima facie obvious before the effective filing date of the claimed invention to have configured the modified Wang 2015 in view of Ishihara to calculate a mean of a first N detected discrete fluorescence values and to set that mean as the first central tendency. This modification would have been possible because the modified Wang 2015 already receives fluorescence data values sequentially during acquisition and organizes them into datasets suitable for averaging, and Ishihara teaches calculating and using mean values as reference central tendencies in fluorescence image processing. Selecting the first N detected values as the initialization set from which a mean is calculated represents a routine software implementation choice that does not alter catheter structure, acquisition timing, or the underlying fluorescence imaging and processing framework. The benefit of the combination is improved stability and predictability in early-stage processing by establishing a defined initial central tendency derived from acquired fluorescence data, thereby supporting consistent subsequent comparison and thresholding operations.
Regarding claim 17, the modified Wang 2015 does not fully teach that the processor: (i) sets the mean of the first N detected discrete fluorescence values as the pending central tendency for a first frame of a pullback procedure, (ii) compares the mean of the first N detected discrete fluorescence values with a second mean of a second N detected discrete fluorescence values to determine whether the second mean is lower than the mean, and (iii) in a case where the second mean is lower than the mean, sets the second mean as the pending central tendency. Rather, as established in claim 16, the modified Wang 2015 teaches receiving a first N detected discrete fluorescence values and calculating a mean of detected fluorescence values within a sequential, frame-based fluorescence processing framework. In particular, as established in the claim 11/16 combination, Ishihara teaches calculating a mean (average gradation value m) for an initial frame (Ishihara, ¶[0083]) and for subsequent frames (Ishihara, ¶[0088]), thereby providing the per-frame mean calculations used in the present limitation. However, the modified Wang 2015 does not explicitly teach the specific comparison-and-update rule in which an initial mean is set as a pending central tendency, a subsequent mean is calculated from a second set of detected fluorescence values, and the pending central tendency is updated when the subsequent mean is lower.
Ishihara teaches calculating and storing a mean for an initial frame (Ishihara, ¶[0083]) and calculating a mean for a subsequent frame (Ishihara, ¶[0088]), thereby establishing that sequential frame-level means are computed and available for comparison. Ma teaches that lower (minimum) intensity values over time are representative of background conditions (Ma, Abstract; Fig. 3), thereby providing a rationale that lower-valued statistical measures (e.g., lower means) better represent background conditions. Lemire teaches that determining a minimum over a sequence of values is performed via sequential comparisons with update when a lower value is identified, stating that “Computing either the global maximum or the global minimum of an array of n elements requires n − 1 comparisons” (Lemire, p. 1) and further teaching streaming/online minimum computation (Lemire, Table I), thereby evidencing an iterative process of initializing a value, comparing subsequent values, and updating when a lower value is found.
These teachings collectively establish a processing paradigm in which (i) an initial mean is set as a pending central tendency, (ii) a subsequent mean is computed and compared to the pending value, and (iii) the pending central tendency is updated when the subsequent mean is lower, directly corresponding to the claimed comparison-and-update logic.
It would have been prima facie obvious before the effective filing date of the claimed invention to have further modified the modified Wang 2015 in view of Ishihara, Ma, and Lemire to set the mean of the first N detected discrete fluorescence values as a pending central tendency, compare that mean with a subsequently calculated mean from a second set of detected fluorescence values, and update the pending central tendency when the subsequent mean is lower. This modification would have been possible because the modified Wang 2015 already processes fluorescence data sequentially during catheter pullback and calculates mean values for frames of acquired data, Ishihara teaches calculating such per-frame means, Ma teaches that lower-valued statistics better represent background conditions, and Lemire teaches the algorithmic framework of sequential comparison with update upon identification of a lower value. The benefit of the combination is improved robustness and accuracy of central tendency tracking during pullback by ensuring that the operative central tendency reflects the lowest observed mean value, thereby supporting reliable background determination and consistent downstream processing.
Regarding claim 18, the modified Wang 2015 does not fully teach that the processor: (i) sets the mean of the first N detected discrete fluorescence values as the pending central tendency for all frames of an entire pullback procedure, (ii) compares the mean of the first N detected discrete fluorescence values with a second mean of a second N detected discrete fluorescence values to determine whether the second mean is lower than the mean, and (iii) in a case where the second mean is lower than the mean, sets the second mean as the pending central tendency. Rather, as established in claim 16, the modified Wang 2015 teaches receiving a first N detected discrete fluorescence values and calculating a mean of detected fluorescence values within a sequential, frame-based fluorescence processing framework. In particular, as established in the claim 11/16 combination, Ishihara teaches calculating a mean (average gradation value m) for an initial frame (Ishihara, ¶[0083]) and for subsequent frames (Ishihara, ¶[0088]), thereby providing the per-frame mean calculations used in the present limitation, and as further established in claim 11, the pending central tendency is maintained across frames until the scanning is finished. However, the modified Wang 2015 does not explicitly teach the specific comparison-and-update rule in which that initial mean is compared against subsequently calculated means and updated when a lower mean is identified.
Ishihara teaches calculating and storing a mean for an initial frame (Ishihara, ¶[0083]) and calculating a mean for a subsequent frame (Ishihara, ¶[0088]), thereby establishing that sequential frame-level means are computed and available for comparison. Ma teaches that lower (minimum) intensity values over time are representative of background conditions (Ma, Abstract; Fig. 3), thereby providing a rationale that lower-valued statistical measures (e.g., lower means) better represent background conditions. Lemire teaches that determining a minimum over a sequence of values is performed via sequential comparisons with update when a lower value is identified, stating that “Computing either the global maximum or the global minimum of an array of n elements requires n − 1 comparisons” (Lemire, p. 1) and further teaching streaming/online minimum computation (Lemire, Table I), thereby evidencing an iterative process of initializing a value, comparing subsequent values, and updating when a lower value is found.
These teachings collectively establish a processing paradigm in which (i) an initial mean is set as a pending central tendency and applied as the operative reference across frames, (ii) a subsequent mean is computed and compared to the pending value, and (iii) the pending central tendency is updated when the subsequent mean is lower, directly corresponding to the claimed comparison-and-update logic applied across all frames of a pullback procedure.
It would have been prima facie obvious before the effective filing date of the claimed invention to have further modified the modified Wang 2015 in view of Ishihara, Ma, and Lemire to set the mean of the first N detected discrete fluorescence values as a pending central tendency for all frames of an entire pullback procedure, compare that mean with subsequently calculated means from additional detected fluorescence values, and update the pending central tendency when a lower mean is identified. This modification would have been possible because the modified Wang 2015 already processes fluorescence data sequentially during catheter pullback and calculates mean values for frames of acquired data, Ishihara teaches calculating such per-frame means, Ma teaches that lower-valued statistics better represent background conditions, and Lemire teaches the algorithmic framework of sequential comparison with update upon identification of a lower value. Applying the pending central tendency across all frames of the pullback represents a continuation of the processing already established in claim 11, where the pending central tendency is maintained until scanning is finished, in combination with Wang 2015’s finite pullback acquisition structure (Wang 2015, p. 7, Sec. 2.2.3) and Lemire’s teaching of zero-latency sequential minimum computation across an entire data stream (Lemire, Table I), thereby establishing that the maintained value persists across all frames from the beginning to the end of the pullback procedure; see also Ughi 2016 (Abstract; p. 1306), describing a complete in vivo pullback imaging procedure with defined start and end, which further supports that the processing spans the entire pullback. The benefit of the combination is improved robustness and consistency of central tendency tracking across the entire pullback by ensuring that the operative central tendency reflects the lowest observed mean value while maintaining a single, consistent reference across frames, thereby supporting reliable background determination and consistent downstream processing.
Regarding claim 19, as shown in claim 11, the modified Wang 2015 does not fully teach that the processor acquires detected fluorescence values of a plurality of A-scan lines that were generated based on detected fluorescence light of the fluorescence collected by the imaging catheter while scanning the bodily lumen in the in vivo environment. Rather, Wang 2015 teaches acquiring detected fluorescence values corresponding to A-scan lines generated during catheter-based scanning, including digitizing fluorescence signals for processing (Wang 2015, p. 6, Sec. 2.2.2: “Signals from the dual balanced detector and the PMT were simultaneously digitized for subsequent processing”), generating fluorescence values on a per A-line basis (Wang 2015, p. 7, Sec. 2.2.3: “The sampling rate of NIRAF signal was 40 kHz, enabling integration of the NIRAF emission within the period of a single A-line”), and mapping fluorescence data points to corresponding OCT A-lines (Wang 2015, p. 8, Sec. 2.2.6: “The normalized NIRAF signal was rendered as a ring shaped image, where each NIRAF data point matches the corresponding OCT A-line at each given rotational position of the catheter”). However, Wang 2015 demonstrates this A-line-based fluorescence acquisition in an ex vivo setting (Sec. 2.2.6), and therefore does not disclose performing the acquisition while scanning the bodily lumen in an in vivo environment.
Ughi 2016 teaches that “The authors present the clinical imaging of human coronary arteries in vivo using a multimodality optical coherence tomography (OCT) and near-infrared autofluorescence (NIRAF) intravascular imaging system and catheter” (Ughi 2016, Abstract) and further teaches that “OCT-NIRAF imaging of the vessel undergoing PCI was performed in all cases” (Ughi 2016, p. 1306), thereby demonstrating that the imaging catheter operates in an in vivo environment during scanning and that detected fluorescence values corresponding to A-scan lines are acquired during in vivo catheter pullback imaging.
It would have been prima facie obvious before the effective filing date of the claimed invention to have further modified the modified Wang 2015 in view of Ughi 2016 to perform the acquisition of detected fluorescence values of a plurality of A-scan lines while scanning the bodily lumen in an in vivo environment with an imaging catheter, because Ughi 2016 explicitly teaches the use of an intravascular OCT-NIRAF imaging catheter in human patients to acquire fluorescence data during in vivo pullback procedures. This modification would have been possible because Wang 2015 already discloses the same OCT-NIRAF catheter architecture, fluorescence detection pathway, and A-line-based data acquisition framework, and Ughi 2016 teaches applying such systems in an in vivo clinical setting. The benefit of the combination is enabling real-time acquisition of fluorescence data corresponding to A-scan lines during in vivo imaging of bodily lumens, thereby allowing clinically relevant fluorescence analysis during catheter-based procedures.
Regarding claim 21, the modified Wang 2015 does not fully teach that the assigning of the central tendency further comprises assigning a lowest central tendency of a portion of the discrete fluorescence data values as the background fluorescence threshold of all of the fluorescence data values where the portion of the discrete fluorescence data values is: (i) less than all of the discrete fluorescence data values, or (ii) less than all of the discrete fluorescence data values and includes a set or predetermined number of lowest fluorescence data values. Rather, as established in claim 1, the modified Wang 2015 teaches acquiring fluorescence data values during catheter-based scanning, calculating central tendencies for frames of acquired data, and assigning a background fluorescence threshold based on central tendency values derived from the acquired data. However, the modified Wang 2015 does not explicitly teach assigning a lowest central tendency derived from only a portion of the discrete fluorescence data values, where the portion is less than all of the data values or corresponds to a subset such as a predetermined number of lowest fluorescence data values, and further applying that resulting threshold across all fluorescence data values. With respect to alternative (i), while Wang 2015 alone does not explicitly teach the portion concept, the combination as applied to claim 1 inherently satisfies this alternative as discussed below.
With respect to alternative (i), the combination as applied to claim 1 already teaches assigning a lowest central tendency derived from frame-level data as the background fluorescence threshold, where each frame represents only a portion of the total fluorescence data values collected during a pullback procedure. Because the fluorescence data values of a single frame constitute less than all fluorescence data values across the dataset, this teaching inherently satisfies assigning a lowest central tendency of a portion less than all of the discrete fluorescence data values as the background fluorescence threshold of all fluorescence data values.
Ma teaches that background estimation may be based on minimum intensity values over time, including determining minimum or lowest-valued intensity data points to represent background conditions (Ma, Abstract; Fig. 3). Ma further teaches segmenting image data into sub-stacks comprising a defined number of frames prior to performing minimum-based background estimation (Ma, Fig. 3), thereby directly teaching selection of a predetermined number of data points as the basis for background estimation. These teachings establish that selecting a subset of lowest-valued data, including a predetermined number of lowest values, provides a robust and noise-resistant estimation of background conditions in fluorescence imaging data.
It would have been prima facie obvious before the effective filing date of the claimed invention to have further modified the modified Wang 2015 in view of Ma to assign the lowest central tendency of a portion of the discrete fluorescence data values as the background fluorescence threshold of all fluorescence data values by deriving the threshold from a selected subset of fluorescence data values. With respect to alternative (i), this behavior is already taught by the combination applied to claim 1, which assigns a lowest central tendency derived from a portion of the data (e.g., frame-level values). With respect to alternative (ii), Ma teaches selecting lowest-valued intensity data points and further teaches performing such selection over defined subsets of data (e.g., sub-stacks of predetermined size), and a person of ordinary skill in the art would have recognized that selecting a predetermined number of such lowest values represents a predictable variation of Ma’s disclosed technique to improve robustness and stability of background estimation. This modification would have been possible because the modified Wang 2015 already processes fluorescence data suitable for central tendency determination, and Ma provides a known technique for selecting lowest-valued data for background estimation. The benefit of the combination is improved robustness and consistency of background fluorescence threshold determination by basing the threshold on representative low-intensity data while ensuring consistent application across the entire dataset, thereby reducing sensitivity to noise and local signal variation.
Claim 20 is rejected under 35 U.S.C. 103 as being unpatentable over Wang et al. (Wang, Hao et al. “Ex Vivo Catheter-Based Imaging of Coronary Atherosclerosis Using Multimodality OCT and NIRAF Excited at 633 Nm.” Biomedical optics express 6.4 (2015): 1363–1375. Web.), hereto referred as Wang 2015, and further in view of Ishihara (US-20120323072-A1), hereto referred as Ishihara, and further in view of Ughi et al. (Ughi, Giovanni J et al. “Clinical Characterization of Coronary Atherosclerosis With Dual-Modality OCT and Near-Infrared Autofluorescence Imaging.” JACC. Cardiovascular imaging 9.11 (2016): 1304–1314. Web.), hereto referred as Ughi 2016, and further in view of Ma et al. (US-20210116380-A1), hereto referred as Ma, and further in view of Lemire (Lemire, Daniel. “Streaming Maximum-Minimum Filter Using No More than Three Comparisons per Element.” ArXiv abs/cs/0610046 (2007)), hereto referred as Lemire, and further in view of Ughi et al. (Ughi et al. “Dual Modality Intravascular Optical Coherence Tomography (OCT) and near-Infrared Fluorescence (NIRF) Imaging: A Fully Automated Algorithm for the Distance-Calibration of NIRF Signal Intensity for Quantitative Molecular Imaging.” International J. Cardiovascular Imaging, 31.2 (2015) 259–268), hereto referred as Ughi 2015.
The modified Wang 2015 teaches claim 11 as shown above.
Regarding claim 20, the modified Wang 2015 does not fully teach that the processor sorts the central tendencies of all frames of the one or more frames from lowest to highest, and assigns the lowest central tendency among the central tendencies of the one or more frames as the background fluorescence threshold for all frames. Rather, as established in claim 11, the modified Wang 2015 teaches acquiring fluorescence data values in a plurality of rotation-synchronized frames during pullback imaging, where each frame corresponds to one catheter rotation and contains discrete fluorescence data values suitable for frame-level statistical processing (Wang 2015, p. 7, Sec. 2.2.3). As further incorporated via Ishihara, the modified Wang 2015 teaches calculating a central tendency for each frame by calculating an average gradation value m of the entire image for each frame and storing such per-frame values for subsequent processing and comparison (Ishihara, ¶[0083]; ¶[0088]). However, while the modified Wang 2015 supports per-frame central tendency calculation and storage, it does not explicitly teach sorting the central tendencies of all frames across an entire pullback from lowest to highest and selecting the lowest central tendency as a single background fluorescence threshold applied to all frames.
Ma teaches that lower (minimum) intensity values over time are representative of background conditions (Ma, Abstract; Fig. 3), thereby providing a rationale that lower-valued statistical measures correspond to background fluorescence levels.
Lemire teaches that determining a minimum over a sequence of values is performed via sequential comparisons with update when a lower value is identified, stating that “Computing either the global maximum or the global minimum of an array of n elements requires n − 1 comparisons” (Lemire, p. 1), and further teaches streaming minimum computation (Lemire, Table I), thereby evidencing that identifying the lowest value across a dataset is a routine operation that may be implemented via comparison or sorting.
Ughi 2015 teaches full-pullback dataset processing in which quantitative analysis is performed using the entire pullback dataset as an input rather than on a frame-by-frame isolated basis, teaching that “the algorithm receives two inputs: the entire IVOCT pullback and the co-registered NIRF dataset” (Ughi 2015, Sec. 2.2), thereby evidencing that identifying and evaluating values across an entire pullback dataset is a routine analysis step in catheter-based imaging workflows. Ughi 2015 further demonstrates that minimum-type metrics may be identified across a pullback dataset (Ughi 2015, Sec. 6), providing additional support that selection of a lowest value across frames is a known analysis approach.
These teachings collectively establish that once per-frame central tendencies are computed and available, it is a routine data processing operation to evaluate those values across the entire pullback and identify the lowest value. A person of ordinary skill in the art would have recognized that sorting a set of already-computed values from lowest to highest is an equivalent and routine alternative to performing sequential minimum-selection (as taught by Lemire) for identifying the lowest value, and that choosing between sorting and running-minimum comparison represents a standard implementation decision.
It would have been prima facie obvious before the effective filing date of the claimed invention to have further modified the modified Wang 2015 in view of Ishihara, Ma, Lemire, and Ughi 2015 to sort the central tendencies of all frames from lowest to highest and assign the lowest central tendency as the background fluorescence threshold applied to all frames. This modification would have been possible because, as already established in claim 11, the processing spans all frames until scanning is finished, and the modified Wang 2015 already calculates and stores per-frame central tendency values during pullback, Ishihara teaches such per-frame mean calculation and storage, Ma teaches that lower-valued statistics represent background conditions, and Lemire teaches efficient minimum-selection across sequences of values with streaming operation across an entire dataset. Ughi 2015 further confirms evaluating metrics across an entire pullback dataset. Sorting represents a straightforward implementation choice for selecting the minimum value among computed central tendencies, and because sorting all frames requires that all frame central tendencies be available, this operation is necessarily performed once scanning has finished, consistent with the claim 11 combination in which processing continues until scanning is finished. The benefit of the combination is improved pullback-wide consistency in background fluorescence normalization by anchoring the background threshold to the lowest central tendency across all frames, thereby reducing sensitivity to localized elevated fluorescence and improving stability and comparability of fluorescence measurements across the entire pullback.
Response to Arguments
Claim Interpretation
Applicant's arguments filed 4/14/2026, pages 12-13, regarding the previous Claim Interpretation of claims 1 and 11 have been fully considered but are not persuasive as shown below.
Applicant’s Argument: Applicant argues that the Examiner’s claim interpretation is unreasonably broad and improperly characterizes “discrete fluorescence data values” as merely frame-based groupings, allegedly importing limitations inconsistent with the specification. Applicant contends that fluorescence data may be grouped into frames regardless of whether such frames are separately collected or contemporaneously acquired during the same scanning procedure.
Examiner’s Response: The argument has been fully considered but is not persuasive. The Examiner applies the broadest reasonable interpretation in light of the specification. The specification teaches that detected fluorescence values are obtained at a rate of one value per A-line and grouped into frames corresponding to a full rotation of the imaging catheter (Instant Application, ¶[0052]: “NIRAF values are collected at a rate of 1 value per A-line and grouped into frames consisting of 500 A-lines… A single frame represents the data collected from one full rotation”). The specification further teaches calculating an average value for each frame (Instant Application, ¶[0052]). Thus, the specification supports that the claimed “discrete fluorescence data values” are distinct A-line-based measurements grouped into rotation-based frames. This interpretation is consistent with that set forth in the prior Office Action, except that the phrasing has been clarified to avoid any implication of temporal separation between frames, as the claims and specification do not require such a limitation.
35 U.S.C. §103
Applicant's arguments filed 4/14/2026, pages 15-39, regarding the previous 103 Rejections of claims 1-20 have been fully considered but are either moot (because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument; that is, there are new grounds of rejection) or are not persuasive as shown below.
Applicant’s Argument: Applicant argues that the cited prior art fails to disclose or suggest performing the claimed scanning, collecting, and calculating steps in an in vivo environment.
Examiner’s Response: To the extent this argument is directed to deficiencies in the prior rejection, the argument is moot in view of the new ground of rejection. The argument is also not persuasive on the merits. Wang 2015 teaches catheter-based fluorescence imaging but performs imaging ex vivo. Ughi 2016 teaches performing OCT-NIRAF imaging in vivo in human patients. It would have been prima facie obvious before the effective filing date of the claimed invention to modify Wang 2015 to perform imaging in an in vivo environment as taught by Ughi 2016 because both references disclose the same type of catheter-based imaging system and Ughi 2016 demonstrates its known clinical implementation. The modification represents a predictable use of a known system for its intended purpose, yielding the benefit of enabling real-time clinical assessment. Indeed, Wang 2015 expressly teaches that the OCT-NIRAF imaging technology is intended for in vivo clinical application (Wang 2015, Abstract), such that the modification to in vivo use realizes the reference’s stated purpose rather than changing it.
Applicant’s argument that the combination changes the principle of operation of Wang 2015 is not persuasive. The modification does not alter the fundamental operation of Wang 2015’s catheter-based OCT-NIRAF imaging system, which remains directed to acquiring fluorescence data during catheter-based imaging. The OCT and NIRAF signal collection, catheter pullback mechanics, and frame acquisition pipeline are preserved; the modification adds processing within the existing data pipeline. The modification merely incorporates known data processing techniques into the existing pipeline and does not render the reference inoperable for its intended purpose. Such modification is consistent with MPEP § 2143.01.
Applicant’s Argument: Applicant argues that the cited prior art fails to disclose or suggest performing collecting and calculating during scanning.
Examiner’s Response: To the extent this argument is directed to deficiencies in the prior rejection, the argument is moot in view of the new ground of rejection. The argument is also not persuasive. Wang 2015 teaches that fluorescence data is acquired continuously during catheter rotation and organized into frames, demonstrating collection during scanning (Wang 2015, p. 7, Sec. 2.2.3). Ishihara teaches processing on a per-frame basis as frames are generated, including calculating average values for each frame and displaying updated images. In view of Ughi 2016’s in vivo pullback imaging, the combination teaches or at least suggests performing processing contemporaneously with scanning. Even under Applicant’s narrower interpretation, it would have been prima facie obvious to apply Ishihara’s per-frame processing to Wang 2015 in view of Ughi 2016 to enable real-time or near real-time evaluation during scanning, a predictable application of known techniques.
Additionally, Lemire teaches minimum determination with zero stream latency (Lemire, Table I), meaning that processing may proceed as each data point is acquired without requiring future data, thereby directly supporting that the claimed operations are performable while scanning is being performed.
Applicant’s Argument: Applicant argues that the cited prior art fails to disclose or suggest calculating a central tendency of the discrete fluorescence data values in each frame while scanning.
Examiner’s Response: The argument has been considered but is not persuasive. While Wang 2015 does not explicitly disclose calculating a central tendency, Ishihara teaches calculating an average value for each frame, which is a central tendency of pixel intensity values. Under the broadest reasonable interpretation, such pixel intensity values correspond to fluorescence data values. As noted above, Ishihara’s gradation values are normalized fluorescence emission intensities derived from fluorescence data via division by a white-light reference image (Ishihara, ¶[0046]). The present specification also confirms central tendency calculations on frame-based fluorescence data (Instant Application, ¶[0052]; ¶[0012]).
To the extent Applicant’s arguments are directed to Ishihara’s FIG. 8 embodiment, such arguments are not persuasive because the present rejection relies on the FIG. 7 embodiment (Ishihara, ¶[0083], ¶[0088]). In the FIG. 7 process, the average value is calculated upon acquisition of each frame (initial (¶[0083]) and each subsequent frame (¶[0088])) without any percentage-change gating condition, thereby teaching calculation of a central tendency for each of the one or more frames as claimed.
Even under Applicant’s narrower interpretation, it would have been prima facie obvious to incorporate such central tendency calculations into the modified Wang 2015 system because applying statistical measures such as averages to measured signal data is a routine and predictable technique that improves signal characterization and reduces noise.
Applicant’s Argument: Applicant argues that the cited references, individually and in combination, contain deficiencies and cannot disclose or suggest the claimed features.
Examiner’s Response: The argument has been considered but is not persuasive. The rejection relies on a combination of references. Wang 2015 teaches catheter-based fluorescence data acquisition and frame-based grouping; Ughi 2016 teaches in vivo implementation; Ishihara teaches per-frame processing and central tendency calculations; and Ma and Lemire provide additional teachings regarding background determination and streaming processing. To the extent Applicant’s arguments are directed to alleged deficiencies in any single reference, such arguments are not persuasive because the rejection relies on the combined teachings.
To the extent Applicant’s arguments are directed to Ughi 2016’s normalization based on minimum and maximum values, such arguments are not persuasive because the present rejection does not rely on Ughi 2016’s normalization teaching for the central tendency limitation, and instead relies on Ishihara and Ma for such teachings. To the extent Applicant argues that Ishihara does not teach determining a lowest central tendency, such argument is not persuasive because the present rejection relies on Ma for minimum-based background estimation and on Lemire for the sequential comparison and update logic for this limitation.
Further, Ma teaches determining background values based on minimum intensity values, and Lemire teaches sequential comparison and updating within streaming data, thereby further supporting that the claimed subject matter would have been obvious.
The rejection is based on articulated reasoning with rational underpinning and does not rely on impermissible hindsight.
Additionally, the cited references teach adjusting fluorescence data values and generating images based on such adjusted values (see, e.g., Wang 2015, Sec. 2.2.5; Ishihara, ¶[0084]).
Accordingly, the claimed subject matter would have been obvious before the effective filing date of the claimed invention.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the
examiner should be directed to AARON MERRIAM whose telephone number is (703) 756-
5938. The examiner can normally be reached M-F 8:00 am - 5:00 pm.
Examiner interviews are available via telephone, in-person, and video conferencing
using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is
encouraged to use the USPTO Automated Interview Request (AIR) at
http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s
supervisor, Jason Sims can be reached on (571)272-4867. The fax phone number for the
organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained
from Patent Center. Unpublished application information in Patent Center is available to
registered users. To file and manage patent submissions in Patent Center, visit:
https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for
more information about Patent Center and https://www.uspto.gov/patents/docx for
information about filing in DOCX format. For additional questions, contact the Electronic
Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO
Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/AARON MERRIAM/Examiner, Art Unit 3791
/MATTHEW KREMER/Primary Examiner, Art Unit 3791
Prosecution Insights