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 .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 4/14/2026 has been entered.
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 1-12, 14-22, 24, and 26-32 are pending. Claims 12, 15, 17, 18, 19, 21, 22, 24, 28, and 30 have been amended. Claims 13, 23, and 25 have been canceled. Claims 1-11, 27, and 29 were previously withdrawn. Claims 31 and 32 are newly added. Claims 12, 14-22, 26, 28, and 30-32 are hereby under examination.
Claim Objections
Claims 14 and 32 are objected to because of the following informalities:
In claim 14, line 5 and line 8 respectively: “the moving image of the first partial area” and “the moving image of the second partial area” do not have proper antecedent basis to claim 12 which only introduces “a first partial moving image” and “a second partial moving image”; and
In claim 32, line 7: “-wherein” should be corrected to “wherein”.
Appropriate correction is required.
Claim Rejections - 35 USC § 103
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 12, 14, 16, 19-20, 28, and 30-32 are rejected under 35 U.S.C. 103 as being unpatentable over Hiroshi et al. (WO-2020138136-A1), hereto referred as Hiroshi, and further in view of Osada et al. (JP-2007117719-A), hereto referred as Osada, and further in view of Hofmann (US-20160296193-A1), hereto referred as Hofmann, and further in view of Bergne et al. (Bergner, Frank, et al. “Autoadaptive Phase‐correlated (AAPC) Reconstruction for 4D CBCT.” Medical Physics [United States], vol. 36, no. 12, December 2009, pp. 5695–706), hereto referred as Bergne.
Regarding claim 12, Hiroshi teaches that an information processing apparatus comprising: at least one memory storing a program; and at least one processor (Hiroshi, FIG. 1, ¶[0020]: "Each of the components of the image processing device 10 described above functions in accordance with a computer program", this explains that the apparatus operates via a program; ¶[0020]: "the CPU uses the RAM as a work area to read and execute a computer program stored in the ROM or a storage unit, thereby realizing the functions of each component", this shows a processor executing a program stored in memory to realize the device functions).
Also regarding claim 12, Hiroshi does not fully teach that the at least one processor, by executing the program, causes the information processing apparatus to acquire projection data obtained by dividing a subject showing a cyclic movement into a first divided area and a second divided area and capturing the first divided area and the second divided area, the projection data including (i) first projection data obtained by capturing the cyclic movement of the subject in a first capturing range including the first divided area for a first time range and (ii) second projection data obtained by capturing the cyclic movement of the subject in a second capturing range including the second divided area for a second time range, wherein the first capturing range and the second capturing range overlap each other in an overlap area. Rather, Hiroshi teaches that "the lungs are divided in the craniocaudal direction so that at least a partial region of the lungs overlap, and a first moving image and a second moving image are acquired" (Hiroshi, ¶[0024], this shows the subject is divided into two areas with an overlap and two corresponding datasets are acquired), and that "the data acquisition unit 110 acquires a first moving image and a second moving image obtained by capturing images of an object from different positions" (Hiroshi, ¶[0024], this teaches acquisition of two datasets captured from different positions). Hiroshi further teaches "three-dimensional tomographic images of multiple time phases obtained by previously imaging different imaging areas of the same subject using the same modality" (Hiroshi, ¶[0017], this shows the first and second datasets correspond to different capturing ranges of the same subject's cyclic movement and respective time ranges). Hiroshi also teaches acquiring phase information of periodic motion of the target organ from the first and second moving images (Hiroshi, ¶¶[0028]-[0032], this shows the subject movement is cyclic or periodic). However, Hiroshi does not expressly use the term "projection data."
Osada teaches that CT acquisition produces raw data that is referred to as projection data, stating: "The pre-processed pure raw data is generally referred to as raw data. Here, the pure raw data and raw data are collectively referred to as 'projection data." (Osada, ¶[0020]). Osada further teaches using stored projection data for reconstruction, stating: "The image reconstruction processing unit 206 performs electrocardiogram-synchronized reconstruction... based on the electrocardiogram signal... and projection data stored in the storage unit 203." (Osada, ¶[0023]). Osada further teaches that projection data is collected and stored with electrocardiogram data, and that, after scanning, the image reconstruction processing unit reconstructs tomographic images using projection data related to the cardiac phase to be reconstructed (Osada, ¶¶[0085]-[0087]). In other words, Osada provides the explicit identification of the CT acquisition data as projection data and its storage and use for reconstruction, which corresponds to the underlying acquisition data necessarily used to produce Hiroshi's tomographic moving images.
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 Hiroshi in view of Osada such that the data acquired to form Hiroshi's CT-based moving images is explicitly characterized as projection data, including first projection data for the first capturing range and second projection data for the second capturing range. The combination would have been feasible because Hiroshi already acquires CT-based moving images for different imaging areas and time phases, and Osada expressly explains that the CT acquisition data used for reconstruction is "projection data" and is stored and used in reconstruction processing. Applying Osada's projection-data framework to Hiroshi's CT acquisition is a routine and predictable implementation choice that clarifies the type of acquired data underlying the reconstructed CT images without requiring any change to the scanning hardware. The benefit of the combination would have been to make explicit that the acquired data for reconstructing the moving images is projection data and to align the acquisition terminology with standard CT reconstruction workflows, thereby improving clarity, reconstruction control, and reproducibility of the imaging pipeline.
Also regarding claim 12, the modified Hiroshi does not fully teach acquire a first partial moving image obtained by reconstructing images of the overlap area of the first capturing range from the first projection data, and a second partial moving image obtained by reconstructing images of the overlap area of the second capturing range from the second projection data. Specifically, the modified Hiroshi teaches acquiring first projection data and second projection data corresponding to first and second CT-based moving images of different imaging areas, as discussed above. Hiroshi teaches acquiring multiple time phase tomographic images for different imaging areas of the same subject and using them as a first moving image and a second moving image (Hiroshi, ¶[0017]: "three-dimensional tomographic images of multiple time phases obtained by previously imaging different imaging areas of the same subject using the same modality"; ¶[0024]: "a first moving image and a second moving image are acquired"). Hiroshi further teaches an overlap area between the first and second moving images (Hiroshi, ¶[0055]: "Region 450 is an area that is only imaged in the first time phase image, region 460 is an area that is only imaged in both the first and second time phase images, and region 470 is an area that is only imaged in the second time phase image"). Osada teaches reconstructing images from projection data, as discussed above (Osada, ¶¶[0020], [0023], [0085]-[0087]). However, the modified Hiroshi does not expressly disclose separately reconstructing images of the overlap area from the first and second projection data as first and second partial moving images for use in the later matching and timing determination.
Hofmann teaches dynamic CT imaging in which projection measurement data is captured for a region of an examination object to be imaged with simultaneous correlated capture of respiratory movement, a phase of the respiratory movement for which image data is to be reconstructed is selected, and projection measurement data assigned to the selected phase is determined (Hofmann, ¶¶[0009]-[0010]). Hofmann further teaches that transition regions of subregions of the region to be imaged between successive respiratory cycles are reconstructed a number of times based on candidate projection measurement data sets, that the candidate projection measurement data sets corresponding to an optimum reconstruction are determined as target projection measurement data, and that a standard reconstruction is performed using the target projection measurement data for each of the successive respiratory cycles (Hofmann, ¶¶[0011]-[0014]). Hofmann also teaches a dynamic CT imaging method comprising "multiple reconstructing transition regions of subregions between successive respiratory cycles" based on candidate projection measurement data sets, determining the candidate projection measurement data sets corresponding to an optimum reconstruction as target projection measurement data, and performing a standard reconstruction using the target projection measurement data (Hofmann, claim 1). Thus, Hofmann teaches reconstructing transition-region image data separately from candidate projection measurement data before performing the later standard reconstruction.
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 Hiroshi in view of Hofmann to acquire the first partial moving image of the overlap area by reconstructing images of the overlap area of the first capturing range from the first projection data and to acquire the second partial moving image of the overlap area by reconstructing images of the overlap area of the second capturing range from the second projection data. The combination would have been feasible because the modified Hiroshi already teaches first and second CT-based moving images reconstructed from projection data and having an overlap/common region between the first and second capturing ranges, and Hofmann teaches multiple reconstruction of transition regions from candidate projection measurement data sets before performing a later standard reconstruction. A person of ordinary skill in the art would have recognized Hiroshi's overlap area between adjacent divided capturing ranges as analogous to Hofmann's transition region between adjacent subregions or respiratory cycles, because both are boundary regions used to ensure that separately reconstructed image portions match when combined or stitched. Hofmann is relied upon for the projection-data-based trial reconstruction and later standard reconstruction technique, not for the particular two-capturing-range geometry, which is taught by Hiroshi. The benefit of the combination would have been to improve reliability of matching and joining divided dynamic CT image ranges by reconstructing and evaluating the overlap or transition region before the larger standard reconstruction, thereby reducing transition-region mismatch, reducing artifacts, and improving the temporal and spatial consistency of the combined dynamic CT image.
Also regarding claim 12, the modified Hiroshi does not fully teach acquire similarity in the cyclic movement of the subject between respective frames of the first partial moving image and the second partial moving image. Specifically, Hiroshi teaches acquiring first and second phase parameters that represent phase information of the periodic motion of the target organ by analyzing the first and second moving images (Hiroshi, ¶¶[0028]-[0032]). Hiroshi further teaches that "the time phase correspondence information acquisition unit 130 associates time phase images of the first moving image and the second moving image that have similar phases based on the first phase parameter and the second phase parameter" and that the time phase correspondence information indicates "which time phase of the first video sequence and which time phase of the second video sequence have the most similar phases" (Hiroshi, ¶[0034]). This teaches acquiring similarity in cyclic movement between respective frames of the first and second moving images. However, Hiroshi does not expressly teach applying the cyclic-movement similarity to separately reconstructed overlap-area partial moving images. Hofmann teaches matching reconstructed transition regions, because Hofmann determines optimum start projection indices for which reconstructed transition regions between successive respiratory cycles match one another best, and calculates the criterion for matching as a displacement value of an artifact metric (Hofmann, claims 3-4). It would have been obvious to apply Hiroshi's phase-similarity matching to the reconstructed overlap or transition-region partial moving images provided by Hofmann, because the modified Hiroshi uses the overlap/common region to align and combine divided capturing ranges, and Hofmann teaches evaluating reconstructed transition regions before the later standard reconstruction to improve matching between image portions. Hiroshi further teaches that image similarity, such as SSD, mutual information, or cross-correlation, may be used in the overlapping region to align corresponding image portions (Hiroshi, ¶[0052]). Hiroshi ¶[0052] is relied upon as support for image comparison in the common region, while Hiroshi ¶[0034] is relied upon as the primary teaching of similarity in cyclic movement. The benefit of the combination would have been to obtain phase correspondence based on the portion of the image data that is actually used for joining the first and second capturing ranges, thereby improving phase matching and reducing mismatch in the combined moving image.
Also regarding claim 12, the modified Hiroshi teaches acquire, by using the similarity, (i) a first timing, within the first time range, for reconstructing images which are used for generating a moving image of the first capturing range from the first projection data, and (ii) a second timing, within the second time range, for reconstructing images which are used for generating a moving image of the second capturing range from the second projection data, wherein the first timing and the second timing correspond to substantially the same phase in the cycle of the cyclic movement of the subject. Specifically, Hiroshi teaches associating time phase images of the first moving image and the second moving image that have similar phases based on the first phase parameter and the second phase parameter (Hiroshi, ¶[0034]: "the time phase correspondence information acquisition unit 130 associates time phase images of the first moving image and the second moving image that have similar phases based on the first phase parameter and the second phase parameter", this teaches selecting corresponding time phases between the first and second moving images based on similarity of phase information). Hiroshi further teaches that the time phase correspondence information indicates which time phase of the first video sequence and which time phase of the second video sequence have the most similar phases (Hiroshi, ¶[0034]). Osada teaches that projection data is stored and used for reconstruction processing (Osada, ¶¶[0020], [0023]). Osada further teaches that projection data is stored together with electrocardiogram data and that, after scanning, the image reconstruction processing unit reconstructs tomographic images using projection data related to the cardiac phase to be reconstructed (Osada, ¶¶[0085]-[0087]). Hofmann teaches that, after phase selection, projection measurement data assigned to the selected phase is determined, and a start projection index interval assigned to the phase projection measurement data may be determined with a plurality of different candidate start projection indices for each start projection index interval (Hofmann, ¶¶[0010], [0024]). Hofmann further teaches multiple reconstructing transition regions of subregions between successive respiratory cycles based on candidate projection measurement data sets, determining the candidate projection measurement data sets that correspond to an optimum reconstruction as target projection measurement data, and performing a standard reconstruction using the target projection measurement data for each of the successive respiratory cycles (Hofmann, ¶¶[0011]-[0014]). Hofmann also teaches determining optimum start projection indices for which the reconstructed transition regions between successive respiratory cycles match one another best (Hofmann, claim 3). Thus, the modified Hiroshi teaches using similarity of cyclic phase and overlap or transition-region matching to determine corresponding reconstruction timings or projection-data intervals for reconstructing images of the first and second capturing ranges from the first and second projection data, wherein the selected timings correspond to substantially the same phase in the cycle of the cyclic movement of the subject.
Also regarding claim 12, the modified Hiroshi does not fully teach wherein a frame rate of the moving image of the first capturing range which is generated by using the images reconstructed from projection data obtained at the first timing, is lower than a frame rate of the first partial moving image. Specifically, the modified Hiroshi teaches acquiring first and second projection data, separately reconstructing overlap or transition-region partial moving images from candidate projection measurement data sets, acquiring similarity between the first and second partial moving images, using overlap or transition-region matching to determine target projection measurement data or corresponding reconstruction timings, and performing a later standard reconstruction using the target projection measurement data. Hofmann further teaches that the optimization process may take place before the final reconstruction and that target projection measurement data forms parts or subsets of the phase projection measurement data assigned to the selected phase (Hofmann, ¶¶[0014]-[0015]). However, the modified Hiroshi does not expressly disclose that the moving image of the first capturing range generated using the images reconstructed from projection data obtained at the first timing has a frame rate lower than the frame rate of the first partial moving image.
Bergner teaches 4D cone-beam CT reconstruction for producing a time series of volumetric images of moving anatomical structures (Bergner, p. 5695, Introduction). Bergner teaches a reconstruction method "combining high temporal resolution inside anatomical regions with strong motion and image quality improvement in regions with little motion" (Bergner, p. 5695, Abstract). Bergner further teaches that, in the proposed method, "the projections are divided into regions that are subject to motion and regions at rest", and that the regions at rest "will be shared among phase bins, leading thus to an overall reduction in artifacts and noise" (Bergner, p. 5695, Abstract). Bergner further teaches that images reconstructed from the method yield "almost the same temporal resolution in the moving volume segments as a conventional phase-correlated reconstruction, while reducing the noise in the motionless regions down to the level of a standard reconstruction without phase correlation" (Bergner, p. 5695, Abstract). Bergner also teaches that conventional phase-correlated reconstruction unnecessarily applies projectionwise phase-dependent weighting to projection regions that are motionless and not in need of such weighting, and that introducing such regions into the reconstruction process reduces noise while keeping temporal resolution high where needed (Bergner, p. 5696). In a phase-binned 4D moving image, the frame rate of the moving image corresponds to the rate or number of distinct temporal phase images presented over the cyclic movement, and therefore sharing image information among phase bins or using fewer distinct phase-dependent reconstructions for a region corresponds to a lower effective frame rate for that region in the moving image. This understanding is consistent with the instant specification, which describes the frame rate of the moving image as the time resolution, explains that a 10 frames/second moving image of the first partial area and a 5 frames/second combined moving image correspond to setting the first timing at a 2-frame interval, and further explains that a user may set the frame rate of the combined moving image to be output and determine a thinning interval so that reconstruction is efficiently performed only at timings sufficient for generating the combined moving image (Instant Application, ¶¶[0023], [0079], [0086]). Thus, Bergner teaches the reconstruction-side principle of using higher temporal resolution for a smaller motion-relevant region while treating the larger or remaining region with lower effective temporal sampling by sharing information among phase bins or by not applying the same phase-correlated reconstruction to regions where the same temporal resolution is not needed.
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 Hiroshi in view of Bergner such that the first partial moving image of the overlap area, which is used for similarity and reconstruction-timing acquisition, is generated at a higher frame rate, while the moving image of the first capturing range generated using images reconstructed from projection data obtained at the first timing is generated at a lower frame rate. The combination would have been feasible because the modified Hiroshi already uses a smaller overlap or transition region for similarity-based matching and reconstruction-timing acquisition before generating the larger capturing-range moving image, and Bergner teaches in 4D CBCT reconstruction that high temporal resolution may be maintained in a smaller motion-relevant region while the remaining region is reconstructed using lower effective temporal sampling by sharing information among phase bins or by not applying the same phase-correlated reconstruction. A person of ordinary skill in the art would have recognized that the overlap or transition region of the modified Hiroshi is the region where dense temporal information is most important, because phase matching accuracy in that region directly affects the selected reconstruction timing and the quality of the combined image. The primary motivation would have been to obtain dense temporal information in the smaller overlap or transition region where phase matching accuracy directly affects reconstruction timing, while avoiding unnecessary high-frame-rate reconstruction of the larger capturing range after the matching timings have already been selected. The benefit of the combination would have been to improve phase matching and reduce motion artifacts in the overlap or transition region while reducing the amount of high-temporal reconstruction performed for the larger capturing range, improving image quality in regions where the same temporal resolution is not needed, and using projection data more efficiently in 4D CT reconstruction. Accordingly, it would have been obvious to generate the moving image of the first capturing range at a lower frame rate than the first partial moving image.
Regarding claim 14, the modified Hiroshi does not fully teach that the at least one processor which, by executing the program, further causes the information processing apparatus to acquire the image of the first divided area, which is reconstructed under a first reconstruction condition at least partially different from a reconstruction condition for reconstructing the moving image of the first partial area, from the first projection data on a basis of the first timing and an image of the second divided area, which is reconstructed under a second reconstruction condition at least partially different from a reconstruction condition for reconstructing the moving image of the second partial area, from the second projection data on a basis of the second timing. Rather, the modified Hiroshi teaches acquiring first and second projection data, acquiring first and second partial moving images of the overlap area from the first and second projection data, acquiring similarity between the first and second partial moving images, and acquiring first and second timings for reconstructing images used for generating moving images of the first and second capturing ranges from the first and second projection data. Specifically, the modified Hiroshi teaches reconstructing the overlap-area partial moving images from the first and second projection data for similarity and timing acquisition, and reconstructing images used for generating moving images of the first and second capturing ranges from the first and second projection data at the first and second timings. However, the modified Hiroshi does not expressly disclose that the reconstruction conditions used for the first and second divided-area images are at least partially different from the reconstruction conditions used for the moving images of the first and second partial areas.
Osada teaches explicit reconstruction from projection data under selectable and differing reconstruction conditions: "The image reconstruction processing unit 206 performs electrocardiogram-synchronized reconstruction... based on ... projection data stored in the storage unit 203" and "has a half reconstruction function and a segment reconstruction function" (Osada, ¶[0023]), with concrete differences in required projection data and processing (e.g., "Half reconstruction requires a group of projection data that covers a range of 180 degrees plus a...", and "In the segment reconstruction method... multiple projection data sets... weighted addition... The weight is determined relatively according to the heart rate", Osada, ¶¶[0024]-[0026]), and with operator-configurable reconstruction conditions including "reconstruction method... reconstruction slice thickness, reconstruction interval" and ECG-gated settings (Osada, ¶¶[0038]-[0042], [0056]).
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 Hiroshi in view of Osada to reconstruct the first and second divided-area images from the respective first and second projection data at the first and second timings under reconstruction conditions that differ at least partially from the conditions used to reconstruct the moving images of the first and second partial areas. The combination would have been feasible because the modified Hiroshi already teaches using overlap-area partial moving images for similarity-based timing acquisition and then reconstructing images of larger capturing ranges from projection data at the selected timings, while Osada provides selectable reconstruction modes and parameters that can be set differently for different reconstruction objectives. Hiroshi's overlap-area partial moving images emphasize temporal resolution and matching accuracy for phase tracking, while Osada's selectable reconstruction conditions allow a CT system to use different reconstruction methods, reconstruction slice thicknesses, reconstruction intervals, and ECG-gated settings for different image outputs. Differing reconstruction conditions naturally follow from these different purposes.
The benefit of the combination would be to optimize temporal characteristics in the moving partial-area reconstructions used for phase tracking and timing acquisition while optimizing spatial fidelity, reconstruction range, slice thickness, reconstruction interval, or signal-to-noise in the divided-area reconstructions generated at the selected timings. One of ordinary skill in the art would have been motivated to make this combination to address known phase mismatch and motion artifact problems in dynamic CT by applying reconstruction conditions appropriate to the imaging purpose, namely, motion analysis and timing acquisition for the partial moving images versus diagnostic or output image generation for the divided-area images.
Regarding claim 16, the modified Hiroshi teaches that the at least one processor which, by executing the program, further causes the information processing apparatus to acquire a combined moving image composed of a plurality of combined images in which a plurality of images of the first divided area and a plurality of images of the second divided area are combined together. Specifically, the modified Hiroshi teaches acquiring a plurality of images of the first divided area and a plurality of images of the second divided area from the first and second projection data at the first and second timings, as discussed above. Hiroshi further teaches that "by combining the time phase images of the first and second moving images associated with each other as the same phase, a combined image at that phase is generated" (Hiroshi, ¶[0014]). This teaches generating a combined moving image composed of a plurality of combined images in which corresponding images of the first divided area and images of the second divided area are combined together.
Regarding claim 19, the modified Hiroshi teaches that the cyclic movement of the subject is respiratory movement of the subject (Hiroshi, ¶[0015]: "a three-dimensional moving image (four-dimensional CT image) of the respiratory movement of the lungs captured by an X-ray CT device will be used as an example", this directly teaches that the cyclic movement of the subject being imaged is respiratory movement of the subject).
Regarding claim 20, the modified Hiroshi teaches that the similarity is proximity of a phase within a respiration cycle in the respiratory movement. Specifically, Hiroshi teaches that a four-dimensional CT image of respiratory movement of the lungs is used as an example (Hiroshi, ¶[0015]: "a three-dimensional moving image (four-dimensional CT image) of the respiratory movement of the lungs captured by an X-ray CT device will be used as an example"). Hiroshi further teaches associating time phase images of the first moving image and the second moving image that have similar phases based on phase parameters (Hiroshi, ¶[0034]: "the time phase correspondence information acquisition unit 130 associates time phase images of the first moving image and the second moving image that have similar phases based on the first phase parameter and the second phase parameter"). Thus, because the modified Hiroshi teaches respiratory movement as the cyclic movement of the subject, and teaches determining similarity based on similar phases of that respiratory movement, Hiroshi teaches that the similarity is proximity of a phase within a respiration cycle in the respiratory movement.
Regarding claim 28, Hiroshi teaches that an information processing method causing a computer to execute the steps of (Hiroshi, ¶[0020]: "Each of the components of the image processing device 10 described above functions in accordance with a computer program", this shows Hiroshi implements the claimed method steps using a computer executing a program).
Also regarding claim 28, Hiroshi does not fully teach the step of acquiring projection data obtained by dividing a subject showing a cyclic movement into a first divided area and a second divided area and capturing the first divided area and the second divided area, the projection data including (i) first projection data obtained by capturing the cyclic movement of the subject in a first capturing range including the first divided area for a first time range and (ii) second projection data obtained by capturing the cyclic movement of the subject in a second capturing range including the second divided area for a second time range, wherein the first capturing range and the second capturing range overlap each other in an overlap area. Rather, Hiroshi teaches that "the lungs are divided in the craniocaudal direction so that at least a partial region of the lungs overlap, and a first moving image and a second moving image are acquired" (Hiroshi, ¶[0024], this shows the subject is divided into two areas with an overlap and two corresponding datasets are acquired), and that "the data acquisition unit 110 acquires a first moving image and a second moving image obtained by capturing images of an object from different positions" (Hiroshi, ¶[0024], this teaches acquisition of two datasets captured from different positions). Hiroshi further teaches "three-dimensional tomographic images of multiple time phases obtained by previously imaging different imaging areas of the same subject using the same modality" (Hiroshi, ¶[0017], this shows the first and second datasets correspond to different capturing ranges of the same subject's cyclic movement and respective time ranges). Hiroshi also teaches acquiring phase information of periodic motion of the target organ from the first and second moving images (Hiroshi, ¶¶[0028]-[0032], this shows the subject movement is cyclic or periodic). However, Hiroshi does not expressly use the term "projection data."
Osada teaches that CT acquisition produces raw data that is referred to as projection data, stating: "The pre-processed pure raw data is generally referred to as raw data. Here, the pure raw data and raw data are collectively referred to as 'projection data." (Osada, ¶[0020]). Osada further teaches using stored projection data for reconstruction, stating: "The image reconstruction processing unit 206 performs electrocardiogram-synchronized reconstruction... based on the electrocardiogram signal... and projection data stored in the storage unit 203." (Osada, ¶[0023]). Osada further teaches that projection data is collected and stored with electrocardiogram data, and that, after scanning, the image reconstruction processing unit reconstructs tomographic images using projection data related to the cardiac phase to be reconstructed (Osada, ¶¶[0085]-[0087]). In other words, Osada provides the explicit identification of the CT acquisition data as projection data and its storage and use for reconstruction, which corresponds to the underlying acquisition data necessarily used to produce Hiroshi's tomographic moving images.
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 Hiroshi in view of Osada such that the data acquired to form Hiroshi's CT-based moving images is explicitly characterized as projection data, including first projection data for the first capturing range and second projection data for the second capturing range. The combination would have been feasible because Hiroshi already acquires CT-based moving images for different imaging areas and time phases, and Osada expressly explains that the CT acquisition data used for reconstruction is "projection data" and is stored and used in reconstruction processing. Applying Osada's projection-data framework to Hiroshi's CT acquisition is a routine and predictable implementation choice that clarifies the type of acquired data underlying the reconstructed CT images without requiring any change to the scanning hardware. The benefit of the combination would have been to make explicit that the acquired data for reconstructing the moving images is projection data and to align the acquisition terminology with standard CT reconstruction workflows, thereby improving clarity, reconstruction control, and reproducibility of the imaging pipeline.
Also regarding claim 28, the modified Hiroshi does not fully teach the step of acquiring a first partial moving image obtained by reconstructing images of the overlap area of the first capturing range from the first projection data and a second partial moving image obtained by reconstructing images of the overlap area of the second capturing range from the second projection data. Specifically, the modified Hiroshi teaches acquiring first projection data and second projection data corresponding to first and second CT-based moving images of different imaging areas, as discussed above. Hiroshi teaches acquiring multiple time phase tomographic images for different imaging areas of the same subject and using them as a first moving image and a second moving image (Hiroshi, ¶[0017]: "three-dimensional tomographic images of multiple time phases obtained by previously imaging different imaging areas of the same subject using the same modality"; ¶[0024]: "a first moving image and a second moving image are acquired"). Hiroshi further teaches an overlap area between the first and second moving images (Hiroshi, ¶[0055]: "Region 450 is an area that is only imaged in the first time phase image, region 460 is an area that is only imaged in both the first and second time phase images, and region 470 is an area that is only imaged in the second time phase image"). Osada teaches reconstructing images from projection data, as discussed above (Osada, ¶¶[0020], [0023], [0085]-[0087]). However, the modified Hiroshi does not expressly disclose separately reconstructing images of the overlap area from the first and second projection data as first and second partial moving images for use in the later matching and timing determination.
Hofmann teaches dynamic CT imaging in which projection measurement data is captured for a region of an examination object to be imaged with simultaneous correlated capture of respiratory movement, a phase of the respiratory movement for which image data is to be reconstructed is selected, and projection measurement data assigned to the selected phase is determined (Hofmann, ¶¶[0009]-[0010]). Hofmann further teaches that transition regions of subregions of the region to be imaged between successive respiratory cycles are reconstructed a number of times based on candidate projection measurement data sets, that the candidate projection measurement data sets corresponding to an optimum reconstruction are determined as target projection measurement data, and that a standard reconstruction is performed using the target projection measurement data for each of the successive respiratory cycles (Hofmann, ¶¶[0011]-[0014]). Hofmann also teaches a dynamic CT imaging method comprising "multiple reconstructing transition regions of subregions between successive respiratory cycles" based on candidate projection measurement data sets, determining the candidate projection measurement data sets corresponding to an optimum reconstruction as target projection measurement data, and performing a standard reconstruction using the target projection measurement data (Hofmann, claim 1). Thus, Hofmann teaches reconstructing transition-region image data separately from candidate projection measurement data before performing the later standard reconstruction.
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 Hiroshi to acquire the first partial moving image of the overlap area by reconstructing images of the overlap area of the first capturing range from the first projection data and to acquire the second partial moving image of the overlap area by reconstructing images of the overlap area of the second capturing range from the second projection data. The combination would have been feasible because the modified Hiroshi already teaches first and second CT-based moving images reconstructed from projection data and having an overlap/common region between the first and second capturing ranges, and Hofmann teaches multiple reconstruction of transition regions from candidate projection measurement data sets before performing a later standard reconstruction. A person of ordinary skill in the art would have recognized Hiroshi's overlap area between adjacent divided capturing ranges as analogous to Hofmann's transition region between adjacent subregions or respiratory cycles, because both are boundary regions used to ensure that separately reconstructed image portions match when combined or stitched. Hofmann is relied upon for the projection-data-based trial reconstruction and later standard reconstruction technique, not for the particular two-capturing-range geometry, which is taught by Hiroshi. The benefit of the combination would have been to improve reliability of matching and joining divided dynamic CT image ranges by reconstructing and evaluating the overlap or transition region before the larger standard reconstruction, thereby reducing transition-region mismatch, reducing artifacts, and improving the temporal and spatial consistency of the combined dynamic CT image.
Also regarding claim 28, the modified Hiroshi does not fully teach the step of acquiring similarity in the cyclic movement of the subject between respective frames of the first partial moving image and the second partial moving image. Specifically, Hiroshi teaches acquiring first and second phase parameters that represent phase information of the periodic motion of the target organ by analyzing the first and second moving images (Hiroshi, ¶¶[0028]-[0032]). Hiroshi further teaches that "the time phase correspondence information acquisition unit 130 associates time phase images of the first moving image and the second moving image that have similar phases based on the first phase parameter and the second phase parameter" and that the time phase correspondence information indicates "which time phase of the first video sequence and which time phase of the second video sequence have the most similar phases" (Hiroshi, ¶[0034]). This teaches acquiring similarity in cyclic movement between respective frames of the first and second moving images. However, Hiroshi does not expressly teach applying the cyclic-movement similarity to separately reconstructed overlap-area partial moving images. Hofmann teaches matching reconstructed transition regions, because Hofmann determines optimum start projection indices for which reconstructed transition regions between successive respiratory cycles match one another best, and calculates the criterion for matching as a displacement value of an artifact metric (Hofmann, claims 3-4). It would have been obvious to apply Hiroshi's phase-similarity matching to the reconstructed overlap or transition-region partial moving images provided by Hofmann, because the modified Hiroshi uses the overlap/common region to align and combine divided capturing ranges, and Hofmann teaches evaluating reconstructed transition regions before the later standard reconstruction to improve matching between image portions. Hiroshi further teaches that image similarity, such as SSD, mutual information, or cross-correlation, may be used in the overlapping region to align corresponding image portions (Hiroshi, ¶[0052]). Hiroshi ¶[0052] is relied upon as support for image comparison in the common region, while Hiroshi ¶[0034] is relied upon as the primary teaching of similarity in cyclic movement. The benefit of the combination would have been to obtain phase correspondence based on the portion of the image data that is actually used for joining the first and second capturing ranges, thereby improving phase matching and reducing mismatch in the combined moving image.
Also regarding claim 28, the modified Hiroshi teaches acquiring, by using the similarity, (i) a first timing, within the first time range, for reconstructing images which are used for generating a moving image of the first capturing range from the first projection data, and (ii) a second timing, within the second time range, for reconstructing images which are used for generating a moving image of the second capturing range from the second projection data, wherein the first timing and the second timing correspond to substantially the same phase in the cycle of the cyclic movement of the subject. Specifically, Hiroshi teaches associating time phase images of the first moving image and the second moving image that have similar phases based on the first phase parameter and the second phase parameter (Hiroshi, ¶[0034]: "the time phase correspondence information acquisition unit 130 associates time phase images of the first moving image and the second moving image that have similar phases based on the first phase parameter and the second phase parameter", this teaches selecting corresponding time phases between the first and second moving images based on similarity of phase information). Hiroshi further teaches that the time phase correspondence information indicates which time phase of the first video sequence and which time phase of the second video sequence have the most similar phases (Hiroshi, ¶[0034]). Osada teaches that projection data is stored and used for reconstruction processing (Osada, ¶¶[0020], [0023]). Osada further teaches that projection data is stored together with electrocardiogram data and that, after scanning, the image reconstruction processing unit reconstructs tomographic images using projection data related to the cardiac phase to be reconstructed (Osada, ¶¶[0085]-[0087]). Hofmann teaches that, after phase selection, projection measurement data assigned to the selected phase is determined, and a start projection index interval assigned to the phase projection measurement data may be determined with a plurality of different candidate start projection indices for each start projection index interval (Hofmann, ¶¶[0010], [0024]). Hofmann further teaches multiple reconstructing transition regions of subregions between successive respiratory cycles based on candidate projection measurement data sets, determining the candidate projection measurement data sets that correspond to an optimum reconstruction as target projection measurement data, and performing a standard reconstruction using the target projection measurement data for each of the successive respiratory cycles (Hofmann, ¶¶[0011]-[0014]). Hofmann also teaches determining optimum start projection indices for which the reconstructed transition regions between successive respiratory cycles match one another best (Hofmann, claim 3). Thus, the modified Hiroshi teaches using similarity of cyclic phase and overlap or transition-region matching to determine corresponding reconstruction timings or projection-data intervals for reconstructing images of the first and second capturing ranges from the first and second projection data, wherein the selected timings correspond to substantially the same phase in the cycle of the cyclic movement of the subject.
Also regarding claim 28, the modified Hiroshi does not fully teach wherein a frame rate of the moving image of the first capturing range which is generated by using the images reconstructed from projection data obtained at the first timing, is lower than a frame rate of the first partial moving image. Specifically, the modified Hiroshi teaches acquiring first and second projection data, separately reconstructing overlap or transition-region partial moving images from candidate projection measurement data sets, acquiring similarity between the first and second partial moving images, using overlap or transition-region matching to determine target projection measurement data or corresponding reconstruction timings, and performing a later standard reconstruction using the target projection measurement data. Hofmann further teaches that the optimization process may take place before the final reconstruction and that target projection measurement data forms parts or subsets of the phase projection measurement data assigned to the selected phase (Hofmann, ¶¶[0014]-[0015]). However, the modified Hiroshi does not expressly disclose that the moving image of the first capturing range generated using the images reconstructed from projection data obtained at the first timing has a frame rate lower than the frame rate of the first partial moving image.
Bergner teaches 4D cone-beam CT reconstruction for producing a time series of volumetric images of moving anatomical structures (Bergner, p. 5695, Introduction). Bergner teaches a reconstruction method "combining high temporal resolution inside anatomical regions with strong motion and image quality improvement in regions with little motion" (Bergner, p. 5695, Abstract). Bergner further teaches that, in the proposed method, "the projections are divided into regions that are subject to motion and regions at rest", and that the regions at rest "will be shared among phase bins, leading thus to an overall reduction in artifacts and noise" (Bergner, p. 5695, Abstract). Bergner further teaches that images reconstructed from the method yield "almost the same temporal resolution in the moving volume segments as a conventional phase-correlated reconstruction, while reducing the noise in the motionless regions down to the level of a standard reconstruction without phase correlation" (Bergner, p. 5695, Abstract). Bergner also teaches that conventional phase-correlated reconstruction unnecessarily applies projectionwise phase-dependent weighting to projection regions that are motionless and not in need of such weighting, and that introducing such regions into the reconstruction process reduces noise while keeping temporal resolution high where needed (Bergner, p. 5696). In a phase-binned 4D moving image, the frame rate of the moving image corresponds to the rate or number of distinct temporal phase images presented over the cyclic movement, and therefore sharing image information among phase bins or using fewer distinct phase-dependent reconstructions for a region corresponds to a lower effective frame rate for that region in the moving image. This understanding is consistent with the instant specification, which describes the frame rate of the moving image as the time resolution, explains that a 10 frames/second moving image of the first partial area and a 5 frames/second combined moving image correspond to setting the first timing at a 2-frame interval, and further explains that a user may set the frame rate of the combined moving image to be output and determine a thinning interval so that reconstruction is efficiently performed only at timings sufficient for generating the combined moving image (Instant Application, ¶¶[0023], [0079], [0086]). Thus, Bergner teaches the reconstruction-side principle of using higher temporal resolution for a smaller motion-relevant region while treating the larger or remaining region with lower effective temporal sampling by sharing information among phase bins or by not applying the same phase-correlated reconstruction to regions where the same temporal resolution is not needed.
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 Hiroshi in view of Bergner such that the first partial moving image of the overlap area, which is used for similarity and reconstruction-timing acquisition, is generated at a higher frame rate, while the moving image of the first capturing range generated using images reconstructed from projection data obtained at the first timing is generated at a lower frame rate. The combination would have been feasible because the modified Hiroshi already uses a smaller overlap or transition region for similarity-based matching and reconstruction-timing acquisition before generating the larger capturing-range moving image, and Bergner teaches in 4D CBCT reconstruction that high temporal resolution may be maintained in a smaller motion-relevant region while the remaining region is reconstructed using lower effective temporal sampling by sharing information among phase bins or by not applying the same phase-correlated reconstruction. A person of ordinary skill in the art would have recognized that the overlap or transition region of the modified Hiroshi is the region where dense temporal information is most important, because phase matching accuracy in that region directly affects the selected reconstruction timing and the quality of the combined image. The primary motivation would have been to obtain dense temporal information in the smaller overlap or transition region where phase matching accuracy directly affects reconstruction timing, while avoiding unnecessary high-frame-rate reconstruction of the larger capturing range after the matching timings have already been selected. The benefit of the combination would have been to improve phase matching and reduce motion artifacts in the overlap or transition region while reducing the amount of high-temporal reconstruction performed for the larger capturing range, improving image quality in regions where the same temporal resolution is not needed, and using projection data more efficiently in 4D CT reconstruction. Accordingly, it would have been obvious to generate the moving image of the first capturing range at a lower frame rate than the first partial moving image.
Regarding claim 30, Hiroshi teaches that a non-transitory computer readable medium that stores a program, wherein the program causes a computer to execute the steps of (Hiroshi, ¶[0020]: "Each of the components of the image processing device 10 described above functions in accordance with a computer program", this shows Hiroshi implements the claimed steps using a computer executing a program; Hiroshi, ¶[0020]: "the CPU uses the RAM as a work area to read and execute a computer program stored in the ROM or a storage unit, thereby realizing the functions of each component", this shows a program stored in a non-transitory storage medium and executed by a computer).
Also regarding claim 30, Hiroshi does not fully teach the step of acquiring projection data obtained by dividing a subject showing a cyclic movement into a first divided area and a second divided area and capturing the first divided area and the second divided area, the projection data including (i) first projection data obtained by capturing the cyclic movement of the subject in a first capturing range including the first divided area for a first time range and (ii) second projection data obtained by capturing the cyclic movement of the subject in a second capturing range including the second divided area for a second time range, wherein the first capturing range and the second capturing range overlap each other in an overlap area. Rather, Hiroshi teaches that "the lungs are divided in the craniocaudal direction so that at least a partial region of the lungs overlap, and a first moving image and a second moving image are acquired" (Hiroshi, ¶[0024], this shows the subject is divided into two areas with an overlap and two corresponding datasets are acquired), and that "the data acquisition unit 110 acquires a first moving image and a second moving image obtained by capturing images of an object from different positions" (Hiroshi, ¶[0024], this teaches acquisition of two datasets captured from different positions). Hiroshi further teaches "three-dimensional tomographic images of multiple time phases obtained by previously imaging different imaging areas of the same subject using the same modality" (Hiroshi, ¶[0017], this shows the first and second datasets correspond to different capturing ranges of the same subject's cyclic movement and respective time ranges). Hiroshi also teaches acquiring phase information of periodic motion of the target organ from the first and second moving images (Hiroshi, ¶¶[0028]-[0032], this shows the subject movement is cyclic or periodic). However, Hiroshi does not expressly use the term "projection data."
Osada teaches that CT acquisition produces raw data that is referred to as projection data, stating: "The pre-processed pure raw data is generally referred to as raw data. Here, the pure raw data and raw data are collectively referred to as 'projection data." (Osada, ¶[0020]). Osada further teaches using stored projection data for reconstruction, stating: "The image reconstruction processing unit 206 performs electrocardiogram-synchronized reconstruction... based on the electrocardiogram signal... and projection data stored in the storage unit 203." (Osada, ¶[0023]). Osada further teaches that projection data is collected and stored with electrocardiogram data, and that, after scanning, the image reconstruction processing unit reconstructs tomographic images using projection data related to the cardiac phase to be reconstructed (Osada, ¶¶[0085]-[0087]). In other words, Osada provides the explicit identification of the CT acquisition data as projection data and its storage and use for reconstruction, which corresponds to the underlying acquisition data necessarily used to produce Hiroshi's tomographic moving images.
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 Hiroshi in view of Osada such that the data acquired to form Hiroshi's CT-based moving images is explicitly characterized as projection data, including first projection data for the first capturing range and second projection data for the second capturing range. The combination would have been feasible because Hiroshi already acquires CT-based moving images for different imaging areas and time phases, and Osada expressly explains that the CT acquisition data used for reconstruction is "projection data" and is stored and used in reconstruction processing. Applying Osada's projection-data framework to Hiroshi's CT acquisition is a routine and predictable implementation choice that clarifies the type of acquired data underlying the reconstructed CT images without requiring any change to the scanning hardware. The benefit of the combination would have been to make explicit that the acquired data for reconstructing the moving images is projection data and to align the acquisition terminology with standard CT reconstruction workflows, thereby improving clarity, reconstruction control, and reproducibility of the imaging pipeline.
Also regarding claim 30, the modified Hiroshi does not fully teach the step of acquiring a first partial moving image obtained by reconstructing images of the overlap area of the first capturing range from the first projection data and a second partial moving image obtained by reconstructing images of the overlap area of the second capturing range from the second projection data. Specifically, the modified Hiroshi teaches acquiring first projection data and second projection data corresponding to first and second CT-based moving images of different imaging areas, as discussed above. Hiroshi teaches acquiring multiple time phase tomographic images for different imaging areas of the same subject and using them as a first moving image and a second moving image (Hiroshi, ¶[0017]: "three-dimensional tomographic images of multiple time phases obtained by previously imaging different imaging areas of the same subject using the same modality"; ¶[0024]: "a first moving image and a second moving image are acquired"). Hiroshi further teaches an overlap area between the first and second moving images (Hiroshi, ¶[0055]: "Region 450 is an area that is only imaged in the first time phase image, region 460 is an area that is only imaged in both the first and second time phase images, and region 470 is an area that is only imaged in the second time phase image"). Osada teaches reconstructing images from projection data, as discussed above (Osada, ¶¶[0020], [0023], [0085]-[0087]). However, the modified Hiroshi does not expressly disclose separately reconstructing images of the overlap area from the first and second projection data as first and second partial moving images for use in the later matching and timing determination.
Hofmann teaches dynamic CT imaging in which projection measurement data is captured for a region of an examination object to be imaged with simultaneous correlated capture of respiratory movement, a phase of the respiratory movement for which image data is to be reconstructed is selected, and projection measurement data assigned to the selected phase is determined (Hofmann, ¶¶[0009]-[0010]). Hofmann further teaches that transition regions of subregions of the region to be imaged between successive respiratory cycles are reconstructed a number of times based on candidate projection measurement data sets, that the candidate projection measurement data sets corresponding to an optimum reconstruction are determined as target projection measurement data, and that a standard reconstruction is performed using the target projection measurement data for each of the successive respiratory cycles (Hofmann, ¶¶[0011]-[0014]). Hofmann also teaches a dynamic CT imaging method comprising "multiple reconstructing transition regions of subregions between successive respiratory cycles" based on candidate projection measurement data sets, determining the candidate projection measurement data sets corresponding to an optimum reconstruction as target projection measurement data, and performing a standard reconstruction using the target projection measurement data (Hofmann, claim 1). Thus, Hofmann teaches reconstructing transition-region image data separately from candidate projection measurement data before performing the later standard reconstruction.
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 Hiroshi in view of Hofmann to acquire the first partial moving image of the overlap area by reconstructing images of the overlap area of the first capturing range from the first projection data and to acquire the second partial moving image of the overlap area by reconstructing images of the overlap area of the second capturing range from the second projection data. The combination would have been feasible because the modified Hiroshi already teaches first and second CT-based moving images reconstructed from projection data and having an overlap/common region between the first and second capturing ranges, and Hofmann teaches multiple reconstruction of transition regions from candidate projection measurement data sets before performing a later standard reconstruction. A person of ordinary skill in the art would have recognized Hiroshi's overlap area between adjacent divided capturing ranges as analogous to Hofmann's transition region between adjacent subregions or respiratory cycles, because both are boundary regions used to ensure that separately reconstructed image portions match when combined or stitched. Hofmann is relied upon for the projection-data-based trial reconstruction and later standard reconstruction technique, not for the particular two-capturing-range geometry, which is taught by Hiroshi. The benefit of the combination would have been to improve reliability of matching and joining divided dynamic CT image ranges by reconstructing and evaluating the overlap or transition region before the larger standard reconstruction, thereby reducing transition-region mismatch, reducing artifacts, and improving the temporal and spatial consistency of the combined dynamic CT image.
Also regarding claim 30, the modified Hiroshi does not fully teach the step of acquiring similarity in the cyclic movement of the subject between respective frames of the first partial moving image and the second partial moving image. Specifically, Hiroshi teaches acquiring first and second phase parameters that represent phase information of the periodic motion of the target organ by analyzing the first and second moving images (Hiroshi, ¶¶[0028]-[0032]). Hiroshi further teaches that "the time phase correspondence information acquisition unit 130 associates time phase images of the first moving image and the second moving image that have similar phases based on the first phase parameter and the second phase parameter" and that the time phase correspondence information indicates "which time phase of the first video sequence and which time phase of the second video sequence have the most similar phases" (Hiroshi, ¶[0034]). This teaches acquiring similarity in cyclic movement between respective frames of the first and second moving images. However, Hiroshi does not expressly teach applying the cyclic-movement similarity to separately reconstructed overlap-area partial moving images. Hofmann teaches matching reconstructed transition regions, because Hofmann determines optimum start projection indices for which reconstructed transition regions between successive respiratory cycles match one another best, and calculates the criterion for matching as a displacement value of an artifact metric (Hofmann, claims 3-4). It would have been obvious to apply Hiroshi's phase-similarity matching to the reconstructed overlap or transition-region partial moving images provided by Hofmann, because the modified Hiroshi uses the overlap/common region to align and combine divided capturing ranges, and Hofmann teaches evaluating reconstructed transition regions before the later standard reconstruction to improve matching between image portions. Hiroshi further teaches that image similarity, such as SSD, mutual information, or cross-correlation, may be used in the overlapping region to align corresponding image portions (Hiroshi, ¶[0052]). Hiroshi ¶[0052] is relied upon as support for image comparison in the common region, while Hiroshi ¶[0034] is relied upon as the primary teaching of similarity in cyclic movement. The benefit of the combination would have been to obtain phase correspondence based on the portion of the image data that is actually used for joining the first and second capturing ranges, thereby improving phase matching and reducing mismatch in the combined moving image.
Also regarding claim 30, the modified Hiroshi teaches acquiring, by using the similarity, (i) a first timing, within the first time range, for reconstructing images which are used for generating a moving image of the first capturing range from the first projection data, and (ii) a second timing, within the second time range, for reconstructing images which are used for generating a moving image of the second capturing range from the second projection data, wherein the first timing and the second timing correspond to substantially the same phase in the cycle of the cyclic movement of the subject. Specifically, Hiroshi teaches associating time phase images of the first moving image and the second moving image that have similar phases based on the first phase parameter and the second phase parameter (Hiroshi, ¶[0034]: "the time phase correspondence information acquisition unit 130 associates time phase images of the first moving image and the second moving image that have similar phases based on the first phase parameter and the second phase parameter", this teaches selecting corresponding time phases between the first and second moving images based on similarity of phase information). Hiroshi further teaches that the time phase correspondence information indicates which time phase of the first video sequence and which time phase of the second video sequence have the most similar phases (Hiroshi, ¶[0034]). Osada teaches that projection data is stored and used for reconstruction processing (Osada, ¶¶[0020], [0023]). Osada further teaches that projection data is stored together with electrocardiogram data and that, after scanning, the image reconstruction processing unit reconstructs tomographic images using projection data related to the cardiac phase to be reconstructed (Osada, ¶¶[0085]-[0087]). Hofmann teaches that, after phase selection, projection measurement data assigned to the selected phase is determined, and a start projection index interval assigned to the phase projection measurement data may be determined with a plurality of different candidate start projection indices for each start projection index interval (Hofmann, ¶¶[0010], [0024]). Hofmann further teaches multiple reconstructing transition regions of subregions between successive respiratory cycles based on candidate projection measurement data sets, determining the candidate projection measurement data sets that correspond to an optimum reconstruction as target projection measurement data, and performing a standard reconstruction using the target projection measurement data for each of the successive respiratory cycles (Hofmann, ¶¶[0011]-[0014]). Hofmann also teaches determining optimum start projection indices for which the reconstructed transition regions between successive respiratory cycles match one another best (Hofmann, claim 3). Thus, the modified Hiroshi teaches using similarity of cyclic phase and overlap or transition-region matching to determine corresponding reconstruction timings or projection-data intervals for reconstructing images of the first and second capturing ranges from the first and second projection data, wherein the selected timings correspond to substantially the same phase in the cycle of the cyclic movement of the subject.
Also regarding claim 30, the modified Hiroshi does not fully teach wherein a frame rate of the moving image of the first capturing range which is generated by using the images reconstructed from projection data obtained at the first timing, is lower than a frame rate of the first partial moving image. Specifically, the modified Hiroshi teaches acquiring first and second projection data, separately reconstructing overlap or transition-region partial moving images from candidate projection measurement data sets, acquiring similarity between the first and second partial moving images, using overlap or transition-region matching to determine target projection measurement data or corresponding reconstruction timings, and performing a later standard reconstruction using the target projection measurement data. Hofmann further teaches that the optimization process may take place before the final reconstruction and that target projection measurement data forms parts or subsets of the phase projection measurement data assigned to the selected phase (Hofmann, ¶¶[0014]-[0015]). However, the modified Hiroshi does not expressly disclose that the moving image of the first capturing range generated using the images reconstructed from projection data obtained at the first timing has a frame rate lower than the frame rate of the first partial moving image.
Bergner teaches 4D cone-beam CT reconstruction for producing a time series of volumetric images of moving anatomical structures (Bergner, p. 5695, Introduction). Bergner teaches a reconstruction method "combining high temporal resolution inside anatomical regions with strong motion and image quality improvement in regions with little motion" (Bergner, p. 5695, Abstract). Bergner further teaches that, in the proposed method, "the projections are divided into regions that are subject to motion and regions at rest", and that the regions at rest "will be shared among phase bins, leading thus to an overall reduction in artifacts and noise" (Bergner, p. 5695, Abstract). Bergner further teaches that images reconstructed from the method yield "almost the same temporal resolution in the moving volume segments as a conventional phase-correlated reconstruction, while reducing the noise in the motionless regions down to the level of a standard reconstruction without phase correlation" (Bergner, p. 5695, Abstract). Bergner also teaches that conventional phase-correlated reconstruction unnecessarily applies projectionwise phase-dependent weighting to projection regions that are motionless and not in need of such weighting, and that introducing such regions into the reconstruction process reduces noise while keeping temporal resolution high where needed (Bergner, p. 5696). In a phase-binned 4D moving image, the frame rate of the moving image corresponds to the rate or number of distinct temporal phase images presented over the cyclic movement, and therefore sharing image information among phase bins or using fewer distinct phase-dependent reconstructions for a region corresponds to a lower effective frame rate for that region in the moving image. This understanding is consistent with the instant specification, which describes the frame rate of the moving image as the time resolution, explains that a 10 frames/second moving image of the first partial area and a 5 frames/second combined moving image correspond to setting the first timing at a 2-frame interval, and further explains that a user may set the frame rate of the combined moving image to be output and determine a thinning interval so that reconstruction is efficiently performed only at timings sufficient for generating the combined moving image (Instant Application, ¶¶[0023], [0079], [0086]). Thus, Bergner teaches the reconstruction-side principle of using higher temporal resolution for a smaller motion-relevant region while treating the larger or remaining region with lower effective temporal sampling by sharing information among phase bins or by not applying the same phase-correlated reconstruction to regions where the same temporal resolution is not needed.
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 Hiroshi in view of Bergner such that the first partial moving image of the overlap area, which is used for similarity and reconstruction-timing acquisition, is generated at a higher frame rate, while the moving image of the first capturing range generated using images reconstructed from projection data obtained at the first timing is generated at a lower frame rate. The combination would have been feasible because the modified Hiroshi already uses a smaller overlap or transition region for similarity-based matching and reconstruction-timing acquisition before generating the larger capturing-range moving image, and Bergner teaches in 4D CBCT reconstruction that high temporal resolution may be maintained in a smaller motion-relevant region while the remaining region is reconstructed using lower effective temporal sampling by sharing information among phase bins or by not applying the same phase-correlated reconstruction. A person of ordinary skill in the art would have recognized that the overlap or transition region of the modified Hiroshi is the region where dense temporal information is most important, because phase matching accuracy in that region directly affects the selected reconstruction timing and the quality of the combined image. The primary motivation would have been to obtain dense temporal information in the smaller overlap or transition region where phase matching accuracy directly affects reconstruction timing, while avoiding unnecessary high-frame-rate reconstruction of the larger capturing range after the matching timings have already been selected. The benefit of the combination would have been to improve phase matching and reduce motion artifacts in the overlap or transition region while reducing the amount of high-temporal reconstruction performed for the larger capturing range, improving image quality in regions where the same temporal resolution is not needed, and using projection data more efficiently in 4D CT reconstruction. Accordingly, it would have been obvious to generate the moving image of the first capturing range at a lower frame rate than the first partial moving image.
Regarding claim 31, the modified Hiroshi does not fully teach that a frame rate of the moving image of the second capturing range which is generated by using the images reconstructed from projection data obtained at the second timing is lower than a frame rate of the second partial moving image. Specifically, the modified Hiroshi teaches acquiring first and second projection data, separately reconstructing overlap or transition-region partial moving images from candidate projection measurement data sets, acquiring similarity between the first and second partial moving images, using overlap or transition-region matching to determine target projection measurement data or corresponding reconstruction timings, and performing a later standard reconstruction using the target projection measurement data. The modified Hiroshi further teaches acquiring the second partial moving image of the overlap area of the second capturing range from the second projection data, and acquiring a second timing, within the second time range, for reconstructing images used for generating a moving image of the second capturing range from the second projection data. Hofmann further teaches that the optimization process may take place before the final reconstruction and that target projection measurement data forms parts or subsets of the phase projection measurement data assigned to the selected phase (Hofmann, ¶¶[0014]-[0015]). However, the modified Hiroshi does not expressly disclose that the moving image of the second capturing range generated using the images reconstructed from projection data obtained at the second timing has a frame rate lower than the frame rate of the second partial moving image.
Bergner teaches 4D cone-beam CT reconstruction for producing a time series of volumetric images of moving anatomical structures (Bergner, p. 5695, Introduction). Bergner teaches a reconstruction method "combining high temporal resolution inside anatomical regions with strong motion and image quality improvement in regions with little motion" (Bergner, p. 5695, Abstract). Bergner further teaches that, in the proposed method, "the projections are divided into regions that are subject to motion and regions at rest", and that the regions at rest "will be shared among phase bins, leading thus to an overall reduction in artifacts and noise" (Bergner, p. 5695, Abstract). Bergner further teaches that images reconstructed from the method yield "almost the same temporal resolution in the moving volume segments as a conventional phase-correlated reconstruction, while reducing the noise in the motionless regions down to the level of a standard reconstruction without phase correlation" (Bergner, p. 5695, Abstract). Bergner also teaches that conventional phase-correlated reconstruction unnecessarily applies projectionwise phase-dependent weighting to projection regions that are motionless and not in need of such weighting, and that introducing such regions into the reconstruction process reduces noise while keeping temporal resolution high where needed (Bergner, p. 5696). In a phase-binned 4D moving image, the frame rate of the moving image corresponds to the rate or number of distinct temporal phase images presented over the cyclic movement, and therefore sharing image information among phase bins or using fewer distinct phase-dependent reconstructions for a region corresponds to a lower effective frame rate for that region in the moving image. This understanding is consistent with the instant specification, which describes the frame rate of the moving image as the time resolution, explains that a 10 frames/second moving image of the first partial area and a 5 frames/second combined moving image correspond to setting the first timing at a 2-frame interval, and further explains that a user may set the frame rate of the combined moving image to be output and determine a thinning interval so that reconstruction is efficiently performed only at timings sufficient for generating the combined moving image (Instant Application, ¶¶[0023], [0079], [0086]). Thus, Bergner teaches the reconstruction-side principle of using higher temporal resolution for a smaller motion-relevant region while treating the larger or remaining region with lower effective temporal sampling by sharing information among phase bins or by not applying the same phase-correlated reconstruction to regions where the same temporal resolution is not needed.
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 Hiroshi in view of Bergner such that the second partial moving image of the overlap area, which is used for similarity and reconstruction-timing acquisition, is generated at a higher frame rate, while the moving image of the second capturing range generated using images reconstructed from projection data obtained at the second timing is generated at a lower frame rate. The combination would have been feasible because the modified Hiroshi already uses a smaller overlap or transition region for similarity-based matching and reconstruction-timing acquisition before generating the larger capturing-range moving image, and Bergner teaches in 4D CBCT reconstruction that high temporal resolution may be maintained in a smaller motion-relevant region while the remaining region is reconstructed using lower effective temporal sampling by sharing information among phase bins or by not applying the same phase-correlated reconstruction. A person of ordinary skill in the art would have recognized that the overlap or transition region of the modified Hiroshi is the region where dense temporal information is most important, because phase matching accuracy in that region directly affects the selected reconstruction timing and the quality of the combined image. The primary motivation would have been to obtain dense temporal information in the smaller overlap or transition region where phase matching accuracy directly affects reconstruction timing, while avoiding unnecessary high-frame-rate reconstruction of the larger second capturing range after the matching timings have already been selected. The benefit of the combination would have been to improve phase matching and reduce motion artifacts in the overlap or transition region while reducing the amount of high-temporal reconstruction performed for the larger second capturing range, improving image quality in regions where the same temporal resolution is not needed, and using projection data more efficiently in 4D CT reconstruction. Accordingly, it would have been obvious to generate the moving image of the second capturing range at a lower frame rate than the second partial moving image.
Regarding claim 32, the modified Hiroshi does not fully teach that a frame rate of the combined moving image is lower than a frame rate of the first partial moving image. Specifically, the modified Hiroshi teaches acquiring first and second projection data, separately reconstructing overlap or transition-region partial moving images from candidate projection measurement data sets, acquiring similarity between the first and second partial moving images, using overlap or transition-region matching to determine target projection measurement data or corresponding reconstruction timings, reconstructing images of the first and second capturing ranges from the first and second projection data at the selected first and second timings, and acquiring a combined moving image by combining the images of the first and second divided areas or capturing ranges. The modified Hiroshi further teaches acquiring the first partial moving image of the overlap area of the first capturing range from the first projection data for similarity-based phase matching and timing acquisition, and acquiring the combined moving image as a later output moving image composed of combined images of the first and second divided areas. Hofmann further teaches that the optimization process may take place before the final reconstruction and that target projection measurement data forms parts or subsets of the phase projection measurement data assigned to the selected phase (Hofmann, ¶¶[0014]-[0015]). However, the modified Hiroshi does not expressly disclose that the frame rate of the combined moving image is lower than the frame rate of the first partial moving image.
Bergner teaches 4D cone-beam CT reconstruction for producing a time series of volumetric images of moving anatomical structures (Bergner, p. 5695, Introduction). Bergner teaches a reconstruction method "combining high temporal resolution inside anatomical regions with strong motion and image quality improvement in regions with little motion" (Bergner, p. 5695, Abstract). Bergner further teaches that, in the proposed method, "the projections are divided into regions that are subject to motion and regions at rest", and that the regions at rest "will be shared among phase bins, leading thus to an overall reduction in artifacts and noise" (Bergner, p. 5695, Abstract). Bergner further teaches that images reconstructed from the method yield "almost the same temporal resolution in the moving volume segments as a conventional phase-correlated reconstruction, while reducing the noise in the motionless regions down to the level of a standard reconstruction without phase correlation" (Bergner, p. 5695, Abstract). Bergner also teaches that conventional phase-correlated reconstruction unnecessarily applies projectionwise phase-dependent weighting to projection regions that are motionless and not in need of such weighting, and that introducing such regions into the reconstruction process reduces noise while keeping temporal resolution high where needed (Bergner, p. 5696). In a phase-binned 4D moving image, the frame rate of the moving image corresponds to the rate or number of distinct temporal phase images presented over the cyclic movement, and therefore sharing image information among phase bins or using fewer distinct phase-dependent reconstructions for a region corresponds to a lower effective frame rate for that region in the moving image. This understanding is consistent with the instant specification, which describes the frame rate of the moving image as the time resolution, explains that a 10 frames/second moving image of the first partial area and a 5 frames/second combined moving image correspond to setting the first timing at a 2-frame interval, and further explains that a user may set the frame rate of the combined moving image to be output and determine a thinning interval so that reconstruction is efficiently performed only at timings sufficient for generating the combined moving image (Instant Application, ¶¶[0023], [0079], [0086]). Thus, Bergner teaches the reconstruction-side principle of using higher temporal resolution for a smaller motion-relevant region while treating the larger or remaining region with lower effective temporal sampling by sharing information among phase bins or by not applying the same phase-correlated reconstruction to regions where the same temporal resolution is not needed.
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 Hiroshi in view of Bergner such that the first partial moving image of the overlap area, which is used for similarity and reconstruction-timing acquisition, is generated at a higher frame rate, while the combined moving image generated from the images reconstructed from the first and second projection data at the selected first and second timings is generated at a lower frame rate. The combination would have been feasible because the modified Hiroshi already uses a smaller overlap or transition region for similarity-based matching and reconstruction-timing acquisition before generating the larger divided-area and combined moving image output, and Bergner teaches in 4D CBCT reconstruction that high temporal resolution may be maintained in a smaller motion-relevant region while the remaining region is reconstructed using lower effective temporal sampling by sharing information among phase bins or by not applying the same phase-correlated reconstruction. A person of ordinary skill in the art would have recognized that the overlap or transition region of the modified Hiroshi is the region where dense temporal information is most important, because phase matching accuracy in that region directly affects the selected reconstruction timing and the quality of the later combined moving image. The primary motivation would have been to obtain dense temporal information in the smaller overlap or transition region where phase matching accuracy directly affects reconstruction timing, while avoiding unnecessary high-frame-rate reconstruction of the larger combined moving image after the matching timings have already been selected. The benefit of the combination would have been to improve phase matching and reduce motion artifacts in the overlap or transition region while reducing the amount of high-temporal reconstruction performed for the larger combined moving image, improving image quality in regions where the same temporal resolution is not needed, and using projection data more efficiently in 4D CT reconstruction. Accordingly, it would have been obvious to generate the combined moving image at a lower frame rate than the first partial moving image.
Claim 15 is rejected under 35 U.S.C. 103 as being unpatentable over Hiroshi et al. (WO-2020138136-A1), hereto referred as Hiroshi, and further in view of Osada et al. (JP-2007117719-A), hereto referred as Osada, and further in view of Hofmann (US-20160296193-A1), hereto referred as Hofmann, and further in view of Bergne et al. (Bergner, Frank, et al. “Autoadaptive Phase‐correlated (AAPC) Reconstruction for 4D CBCT.” Medical Physics [United States], vol. 36, no. 12, December 2009, pp. 5695–706), hereto referred as Bergne and further in view of Tsukagoshi (US-20040190674-A1), hereto referred as Tsukagoshi.
The modified Hiroshi teaches claim 12 as described above.
Regarding claim 15, the modified Hiroshi does not fully teach that the first divided area includes the overlap area and an area other than the overlap area, the first reconstruction condition is a reconstruction condition in which the first divided area is a reconstruction range, the second divided area includes the overlap area and an area other than the overlap area, and the second reconstruction condition is a reconstruction condition in which the second divided area is a reconstruction range. Rather, the modified Hiroshi teaches that the first capturing range and the second capturing range overlap each other in an overlap area, and that first and second partial moving images of the overlap area are acquired from the first and second projection data, as discussed above. Hiroshi further teaches that the divided imaging areas include both an overlap area and areas other than the overlap area, because Hiroshi teaches that "Region 450 is an area that is only imaged in the first time phase image, region 460 is an area that is only imaged in both the first and second time phase images, and region 470 is an area that is only imaged in the second time phase image" (Hiroshi, ¶[0055]). Thus, Hiroshi teaches that the first divided area includes the overlap area and an area other than the overlap area, and that the second divided area includes the overlap area and an area other than the overlap area. However, the modified Hiroshi does not expressly disclose defining the first reconstruction condition as a reconstruction condition in which the first divided area is a reconstruction range, or defining the second reconstruction condition as a reconstruction condition in which the second divided area is a reconstruction range.
Tsukagoshi teaches determining a reconstruction range and determining a scan range covering the reconstruction range (Tsukagoshi, ¶[0036]: "...determines... the reconstruction range 111... and determines the scan range 112... covering the reconstruction range 111"), and reconstructing image data for slices included in the reconstruction range on the basis of projection data acquired by scans (Tsukagoshi, ¶[0041]: "...reconstructs image data... for each of plural slices... of the reconstruction range... on the basis of the projection data acquired by Scans"). Tsukagoshi further teaches that scanning is performed in a scan range corresponding to a reconstruction range and reconstructing image data included in the reconstruction range on the basis of projection data acquired by the scanning (Tsukagoshi, claim 13: "performing scanning in a scan range corresponding to said reconstruction range; and" and "reconstructing image data related to plural slices, parallel to one another and included in said reconstruction range, slice-by-slice on the basis of projection data acquired by said scanning"). These teachings support using a specified reconstruction range as a reconstruction condition, and that the reconstruction range may be set to a desired anatomical coverage within the scan.
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 Hiroshi in view of Tsukagoshi to set the first reconstruction condition such that the first divided area is the reconstruction range, and to set the second reconstruction condition such that the second divided area is the reconstruction range. The combination would have been feasible because the modified Hiroshi already teaches first and second divided areas that each include the overlap area and an area other than the overlap area, and further teaches reconstructing images of the first and second divided areas from the first and second projection data on the basis of the first and second timings, while Tsukagoshi teaches the routine CT configuration of defining a reconstruction range and reconstructing image data included in that reconstruction range on the basis of acquired projection data. Applying Tsukagoshi's reconstruction-range configuration to the modified Hiroshi would have merely involved setting the reconstruction range for each divided-area reconstruction to correspond to the divided area being reconstructed.
The benefit of the combination would have been to ensure that each divided-area image is reconstructed over the full divided area needed for later output or combination, while still allowing the overlap-area partial moving images to be reconstructed under separate reconstruction conditions for phase matching and timing acquisition as discussed above. One of ordinary skill in the art would have been motivated to make this combination to define reconstruction ranges appropriate to the intended image output, thereby improving consistency between the selected projection-data timing, the reconstructed divided-area images, and the later combined moving image.
Claim 17 is rejected under 35 U.S.C. 103 as being unpatentable over Hiroshi et al. (WO-2020138136-A1), hereto referred as Hiroshi, and further in view of Osada et al. (JP-2007117719-A), hereto referred as Osada, and further in view of Hofmann (US-20160296193-A1), hereto referred as Hofmann, and further in view of Bergne et al. (Bergner, Frank, et al. “Autoadaptive Phase‐correlated (AAPC) Reconstruction for 4D CBCT.” Medical Physics [United States], vol. 36, no. 12, December 2009, pp. 5695–706), hereto referred as Bergne, and further in view of Yamazaki (US-20070147576-A1), hereto referred as Yamazaki.
The modified Hiroshi teaches claim 12 as described above.
Regarding claim 17, the modified Hiroshi does not fully teach that the at least one processor which, by executing the program, further causes the information processing apparatus to acquire combined projection data in which the first projection data and the second projection data are combined together by connecting projection data in the first projection data which is obtained at the first timing and projection data in the second projection data which is obtained at the second timing, and acquire a combined moving image by using the combined projection data. Rather, the modified Hiroshi teaches acquiring first and second projection data, acquiring first and second timings corresponding to substantially the same phase in the cycle of the cyclic movement, and generating a combined moving image from images reconstructed from the first and second projection data at the selected timings, as discussed above. Hiroshi further teaches generating a combined image by combining time-phase images of the first and second moving images associated with each other as the same phase (Hiroshi, ¶[0014]; FIG. 2, combined image generating unit 150). However, the modified Hiroshi does not expressly disclose combining the first and second projection data together by connecting projection data obtained at the first timing and projection data obtained at the second timing before acquiring the combined moving image.
Yamazaki teaches an X-ray CT apparatus in which a reconstruction unit is configured to generate compounded projection data based on wide range projection data and narrow range projection data and to perform a reconstruction process based on the compounded projection data (Yamazaki, claim 1). Yamazaki further teaches that a compounding unit generates projection data based on wide range projection data supplied from a wide range projection data memory unit and narrow range projection data supplied from a narrow range projection data memory unit (Yamazaki, claim 4). Yamazaki further teaches that, after wide range projection data is reconstructed into images and converted into projection data passing along the same X-ray paths around edge portions of the narrow range projection data, a data compounding unit compounds the re-generated wide range projection data with the narrow range projection data, and the compounded projection data is processed in a reconstruction processing unit (Yamazaki, ¶¶[0035]-[0036]). Thus, Yamazaki teaches combining or compounding different projection-data sets and then reconstructing using the combined or compounded projection data.
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 Hiroshi in view of Yamazaki to acquire combined projection data by connecting projection data in the first projection data obtained at the first timing and projection data in the second projection data obtained at the second timing, and to acquire the combined moving image by using the combined projection data. Yamazaki is relied upon for the projection-data-level combination followed by reconstruction, not for the particular timing-selection framework, which is supplied by the modified Hiroshi, as modified above. The combination would have been feasible because the modified Hiroshi already teaches selecting corresponding first and second timings for first and second projection data and generating a combined moving image from the corresponding first and second divided-area image information, while Yamazaki teaches the CT reconstruction technique of combining different projection-data sets and performing reconstruction based on the compounded projection data. Applying Yamazaki's projection-data compounding technique to the modified Hiroshi would have merely involved performing the combination at the projection-data stage for corresponding first and second timing data, rather than only combining the already reconstructed image data.
The benefit of the combination would have been to improve consistency of the combined moving image by reconstructing from projection data that has already been connected or compounded at corresponding phases, thereby reducing errors that may arise when separately reconstructed images are combined only at the image level. One of ordinary skill in the art would have been motivated to make this combination to use projection-data-level compounding for a combined CT reconstruction while retaining the phase-matched timing selection supplied by the modified Hiroshi, as modified above.
Claims 18, 24, and 26 are rejected under 35 U.S.C. 103 as being unpatentable over Hiroshi et al. (WO-2020138136-A1), hereto referred as Hiroshi, and further in view of Osada et al. (JP-2007117719-A), hereto referred as Osada, and further in view of Hofmann (US-20160296193-A1), hereto referred as Hofmann, and further in view of Bergne et al. (Bergner, Frank, et al. “Autoadaptive Phase‐correlated (AAPC) Reconstruction for 4D CBCT.” Medical Physics [United States], vol. 36, no. 12, December 2009, pp. 5695–706), hereto referred as Bergne, and further in view of Johnston et al. (US-20120177271-A1), hereto referred as Johnston.
The modified Hiroshi teaches claim 12 as described above.
Regarding claim 18, the modified Hiroshi does not fully teach that the at least one processor which, by executing the program, further causes the information processing apparatus to determine, on a basis of a combination having higher similarity from among combinations of first time positions in the first time range and second time positions in the second time range, the first time positions as the first timing and the second time positions as the second timing. Rather, the modified Hiroshi teaches acquiring similarity in the cyclic movement of the subject between respective frames of the first partial moving image and the second partial moving image, and acquiring, by using the similarity, first and second timings within the first and second time ranges, as discussed above. Hiroshi further teaches associating time phase images of the first moving image and the second moving image that have similar phases based on phase parameters (Hiroshi, ¶[0034]: "the time phase correspondence information acquisition unit 130 associates time phase images of the first moving image and the second moving image that have similar phases based on the first phase parameter and the second phase parameter"), which supports determining corresponding time phases or time positions between the first and second datasets based on similarity. Hiroshi further teaches that, when multiple time phases on one side are most similar to the same time phase on the other side, only one pair of most similar phase parameters may be associated (Hiroshi, ¶[0044]: "When the phase parameters of a plurality of time phases on one side are all most similar to the same time phases on the other side, only one pair of most similar phase parameters may be associated with each other"), which supports selecting a higher-similarity pairing from among candidate pairings. However, the modified Hiroshi does not expressly recite determining the first timing and the second timing on the basis of a higher-similarity combination defined from among combinations of first time positions in the first time range and second time positions in the second time range, as recited.
Johnston teaches selecting, from among multiple candidate images within a bin, the set of images whose similarity is maximized using a correlation coefficient based optimization, where the least cost path represents the set of images whose similarity with adjacent images is maximized (Johnston, ¶[0034]: "Each arc has a cost equal to 1/K where K is the two dimensional correlation coefficient"; ¶[0037]: "The least cost path represents the set of images whose similarity with adjacent images is maximized"). These teachings support selecting, from among combinations of candidate time-position images, the combination having higher similarity.
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 Hiroshi in view of Johnston so that, when determining corresponding time phases between the first and second datasets based on similarity, the processor evaluates combinations of candidate first time positions within the first time range and candidate second time positions within the second time range, selects a combination having higher similarity, and determines the first time positions as the first timing and the second time positions as the second timing on the basis of the selected higher-similarity combination. The combination would have been feasible because Hiroshi already performs similarity-based association between time phase images across two datasets and contemplates selecting only the most similar pairing when multiple candidates exist, Johnston teaches selecting an optimal set by maximizing similarity or correlation from among candidate combinations, and the modified Hiroshi already establishes first and second time ranges and corresponding reconstruction timings for the first and second projection data. Integrating these teachings merely requires applying Johnston's similarity-optimization selection to Hiroshi's candidate time-phase associations within the respective first and second time ranges and then using the selected time positions as the reconstruction timings, which is a predictable software selection step within CT processing. The benefit of the combination would have been to improve temporal correspondence between the first and second divided-area reconstructions by selecting higher-similarity time-position combinations from within the respective first and second time ranges, thereby reducing mismatch and improving image quality.
Regarding claim 24, the modified Hiroshi does not fully teach that the at least one processor which, by executing the program, further causes the information processing apparatus to: calculate the similarity between frame images by moving a relative position of a frame image of the first partial moving image and a frame image of the second partial moving image, and determine, on a basis of combinations having the higher similarity from among combinations of frame images of first time positions in the first time range and frame images of second time positions in the second time range, the first time positions as the first timing and the second time positions as the second timing. Rather, the modified Hiroshi teaches acquiring first and second partial moving images of the overlap area, acquiring similarity in the cyclic movement of the subject between respective frames of the first partial moving image and the second partial moving image, and acquiring first and second timings within the first and second time ranges by using the similarity, as discussed above. Hiroshi further teaches associating time phase images of the first moving image and the second moving image that have similar phases based on phase parameters (Hiroshi, ¶[0034]: "the time phase correspondence information acquisition unit 130 associates time phase images of the first moving image and the second moving image that have similar phases based on the first phase parameter and the second phase parameter"), which supports calculating similarity between frames and determining corresponding time positions based on similarity. Hiroshi further teaches that, when multiple time phases on one side are most similar to the same time phase on the other side, only one pair of most similar phase parameters may be associated (Hiroshi, ¶[0044]: "When the phase parameters of a plurality of time phases on one side are all most similar to the same time phases on the other side, only one pair of most similar phase parameters may be associated with each other"), which supports selecting a higher-similarity pairing from among candidate pairings. Hiroshi further teaches calculating a joining position by searching for a corresponding position between first and second time-phase images in an overlapping region and searching for a slice position at which image similarity between tomographic images is high, using image similarity such as SSD, mutual information, or cross-correlation coefficient (Hiroshi, ¶¶[0051]-[0052]). However, the modified Hiroshi does not expressly disclose calculating the similarity between frame images by moving a relative position of a frame image of the first partial moving image and a frame image of the second partial moving image, nor does Hiroshi expressly describe determining the first and second timings based on combinations of frame images from the first and second time ranges having the higher similarity.
Johnston teaches calculating similarity between images using a correlation coefficient and selecting an optimal set by maximizing similarity among candidate images (Johnston, ¶[0034]: "Each arc has a cost equal to 1/K where K is the two dimensional correlation coefficient"; ¶[0037]: "The least cost path represents the set of images whose similarity with adjacent images is maximized"). Johnston further teaches evaluating similarity while considering positional displacement between images, including selecting an image "at a displacement which is closest" and determining a candidate set of images within a displacement range (Johnston, ¶[0039]: "...select an image in the bin at a displacement which is closest to the current displacement"; ¶[0040]: "...find the set of possible candidate images which lie within a range of the current displacement..."). These teachings support calculating similarity by moving a relative position or displacement between images and selecting higher-similarity candidate combinations.
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 Hiroshi in view of Johnston so that, when calculating similarity between frame images of the first and second partial moving images, the similarity calculation is performed while moving a relative position of a frame image of the first partial moving image and a frame image of the second partial moving image, for example, by evaluating similarity under different displacement offsets applied to one frame image relative to the other, and so that the determining includes selecting, from among combinations of frame images of candidate first time positions within the first time range and frame images of candidate second time positions within the second time range, combinations having the higher similarity, and determining the first and second time positions as the first and second timings based on the higher-similarity combinations. The combination would have been feasible because Hiroshi already performs similarity-based association between time phase images across two datasets and searches for corresponding positions in an overlap region based on image similarity, Johnston teaches computing similarity via correlation while accounting for positional displacement and selecting a higher-similarity candidate pairing, and the modified Hiroshi already establishes first and second partial moving images, first and second time ranges, and reconstruction timings for the first and second projection data. Integrating these teachings merely requires implementing the similarity computation between the corresponding frame images of the first and second partial moving images with a displacement search or relative-position shift as in Johnston and then using the resulting higher-similarity pairing to define the reconstruction timings, which is a predictable software modification.
The benefit of the combination would have been improved robustness of similarity calculation and correspondence determination between the first and second partial moving images by compensating for spatial misalignment through relative movement, thereby improving reconstruction timing selection and reducing artifacts in the later divided-area and combined moving images.
Regarding claim 26, the modified Hiroshi does not fully teach that the at least one processor which, by executing the program, further causes the information processing apparatus to determine, for each combination of the first time positions and the second time positions, a movement amount of the relative position, the movement amount comprising a magnitude of a positional displacement applied to the frame image to calculate similarity; assign a priority ranking to the combinations of the first time positions and the second time positions based on the movement amount; and determine the first time positions as the first timing and the second time positions as the second timing on a basis of the combination having the higher priority from among the combinations having the higher similarity, wherein the higher priority is determined based on the movement amount. Rather, the modified Hiroshi teaches acquiring first and second partial moving images of the overlap area, calculating similarity between frame images by moving a relative position of a frame image of the first partial moving image and a frame image of the second partial moving image, and determining first and second timings based on combinations having higher similarity from among combinations of frame images of first time positions in the first time range and frame images of second time positions in the second time range, as discussed above. However, the modified Hiroshi does not expressly disclose determining, for each combination, a movement amount of the relative position, assigning a priority ranking to the combinations based on the movement amount, and selecting, from among higher-similarity combinations, a higher-priority combination based on the movement amount.
Hofmann teaches determining optimum start projection indices from among a plurality of different candidate start projection indices, wherein the optimum start projection indices are those "for which the reconstructed transition regions between successive respiratory cycles match one another best" and wherein a criterion for how closely the reconstructed transition regions match is calculated "as a displacement value of an artifact metric" (Hofmann, claim 3). Hofmann further teaches that the respective candidate start projection indices "for which the displacement value of the artifact metric is a minimum, are determined as the optimum start projection indices" (Hofmann, claim 4). Thus, Hofmann teaches evaluating candidate timing or projection-index combinations based on a displacement value, and selecting the candidate combination having a more favorable displacement value, such as a minimum displacement value, as the optimum timing or projection-index selection.
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 Hiroshi in view of Hofmann so that, for each combination of first time positions and second time positions, the processor determines a movement amount of the relative position, assigns a priority ranking to the combinations based on the movement amount, and determines the first and second time positions as the first and second timings on the basis of the combination having higher priority from among the combinations having higher similarity, wherein the higher priority is determined based on the movement amount. The combination would have been feasible because the modified Hiroshi already teaches calculating similarity between frame images while moving a relative position and selecting higher-similarity combinations of first and second time positions, and Hofmann teaches evaluating candidate timing or projection-index selections using a displacement value and selecting the optimum candidates based on the displacement value. Applying Hofmann's displacement-based selection criterion to the modified Hiroshi would have merely involved ranking higher-similarity candidate frame combinations according to the amount of positional movement needed to obtain the similarity, and then selecting a higher-priority candidate, such as a candidate requiring a smaller movement amount, from among those higher-similarity combinations.
The benefit of the combination would have been to improve the reliability of timing selection by preferring, among combinations that already have high similarity, combinations that require less positional displacement to achieve the similarity. One of ordinary skill in the art would have been motivated to make this combination because a smaller required relative-position movement indicates better inherent alignment between the first and second partial moving images, thereby reducing the risk that a high similarity score results from excessive positional shifting rather than true phase correspondence, and improving the accuracy of the selected first and second timings.
Claims 21-22 are rejected under 35 U.S.C. 103 as being unpatentable over Hiroshi et al. (WO-2020138136-A1), hereto referred as Hiroshi, and further in view of Osada et al. (JP-2007117719-A), hereto referred as Osada, and further in view of Hofmann (US-20160296193-A1), hereto referred as Hofmann, and further in view of Bergne et al. (Bergner, Frank, et al. “Autoadaptive Phase‐correlated (AAPC) Reconstruction for 4D CBCT.” Medical Physics [United States], vol. 36, no. 12, December 2009, pp. 5695–706), hereto referred as Bergne and further in view of Choi (US-20090066782-A1), hereto referred as Choi.
The modified Hiroshi teaches claim 12 as described above.
Regarding claim 21, the modified Hiroshi does not fully teach that the at least one processor which, by executing the program, further causes the information processing apparatus to determine a frame rate of the first partial moving image according to an observation site of the subject included in the first capturing range. Specifically, the modified Hiroshi teaches acquiring the first partial moving image of the overlap area of the first capturing range from the first projection data, and using the first partial moving image for similarity-based phase matching and reconstruction-timing acquisition, as discussed above. Bergner further teaches 4D cone-beam CT reconstruction in which high temporal resolution is used inside anatomical regions with strong motion and regions at rest are shared among phase bins, as discussed above. However, the modified Hiroshi does not expressly disclose determining the frame rate of the first partial moving image according to an observation site of the subject included in the first capturing range.
Choi teaches an image sensor that provides adaptive spatial-temporal multi-resolution for a specific region of interest, and teaches using one channel at a low frame rate for stationary backgrounds and another channel at a high frame rate for moving objects in a region of interest (Choi, ¶[0004]). Choi further teaches that the high-frame readout occurs substantially only in the region of interest, thereby reducing bandwidth and power consumption (Choi, ¶[0005]). Choi also teaches a first data channel that outputs image data for a background region at a first frame rate and a second data channel that outputs image data for a region of interest at a second frame rate faster than the first frame rate (Choi, claim 2). Thus, Choi teaches determining a frame rate based on the region or observation site being imaged, such that different regions receive different frame rates according to the imaging need for 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 Hiroshi in view of Choi to determine the frame rate of the first partial moving image according to an observation site of the subject included in the first capturing range. The combination would have been feasible because the modified Hiroshi already teaches reconstructing a first partial moving image of an overlap area from first projection data for phase matching and timing acquisition, Bergner teaches in 4D CT reconstruction that temporal resolution may be varied by anatomical region or motion-relevant region, and Choi teaches the general image-processing principle of assigning different frame rates to different regions, including assigning a higher frame rate to a region of interest and a lower frame rate to background. Applying Choi's region-dependent frame-rate selection to the modified Hiroshi would have merely involved setting the frame rate of the first partial moving image based on the observation site included in the first capturing range, such as whether the observation site requires higher temporal resolution for motion tracking, phase matching, or reduced motion blur.
The benefit of the combination would have been to allocate higher frame rate processing to an observation site where temporal information is important for phase matching and reconstruction-timing acquisition, while avoiding unnecessary high-frame-rate processing for regions where the same temporal information is not needed. One of ordinary skill in the art would have been motivated to make this combination to improve phase matching accuracy for the first partial moving image while reducing processing burden, bandwidth, and storage associated with high-frame-rate image generation.
Regarding claim 22, the modified Hiroshi does not fully teach that the at least one processor which, by executing the program, further causes the information processing apparatus to determine a frame rate of the first partial moving image according to a change in the cyclic movement of the subject. Specifically, the modified Hiroshi teaches acquiring the first partial moving image of the overlap area of the first capturing range from the first projection data, and using the first partial moving image for similarity-based phase matching and reconstruction-timing acquisition, as discussed above. Hiroshi further teaches that the subject undergoes periodic respiratory movement and that phase information of the periodic motion is acquired from the moving images (Hiroshi, ¶¶[0015], [0028]-[0032]). Bergner further teaches 4D cone-beam CT reconstruction in which high temporal resolution is used inside anatomical regions with strong motion and regions at rest are shared among phase bins, as discussed above. However, the modified Hiroshi does not expressly disclose determining the frame rate of the first partial moving image according to a change in the cyclic movement of the subject.
Choi teaches an image sensor that provides adaptive spatial-temporal multi-resolution for tracking movement in a region of interest. Choi teaches using one channel at a low frame rate for stationary backgrounds and another channel at a high frame rate for moving objects in a region of interest (Choi, ¶[0004]). Choi further teaches that the high-frame readout occurs substantially only in the region of interest, thereby reducing bandwidth and power consumption (Choi, ¶[0005]). Choi also teaches detecting motion between a current frame and a previous frame using a motion comparator (Choi, Abstract; claim 1), and teaches outputting image data for a background region at a first frame rate and image data for a region of interest at a second frame rate faster than the first frame rate (Choi, claim 2). Thus, Choi teaches the general motion-responsive frame-rate principle that image data corresponding to moving or changing content is assigned a higher frame rate than image data corresponding to stationary or lower-motion content.
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 Hiroshi in view of Choi to determine the frame rate of the first partial moving image according to a change in the cyclic movement of the subject. The combination would have been feasible because the modified Hiroshi already teaches using the first partial moving image of the overlap area for phase matching and timing acquisition in a subject undergoing cyclic movement, Bergner teaches in 4D CT reconstruction that high temporal resolution may be applied to regions with strong motion while regions with little motion are treated with lower effective temporal sampling, and Choi teaches assigning different frame rates based on whether image content is moving or stationary. Although Choi assigns frame rate based on motion-responsive region selection, one of ordinary skill in the art would have recognized that the same underlying temporal-sampling principle applies to the speed or amount of change in a cyclic movement over time, because faster or larger changes in the cyclic movement require denser temporal information to identify corresponding phases accurately, while slower or smaller changes require less dense temporal information.
The benefit of the combination would have been to provide a higher frame rate for the first partial moving image during time ranges in which changes in the cyclic movement make dense temporal information important for phase matching, while avoiding unnecessary high-frame-rate reconstruction during time ranges in which the cyclic movement changes less. One of ordinary skill in the art would have been motivated to make this combination to improve phase matching and timing accuracy for the overlap-area partial moving image while using reconstruction resources efficiently.
Response to Arguments
Objections
Applicant's arguments filed 4/14/2026, page 18, regarding the previous Objections of claims 24, 28, and 30 have been fully considered and are persuasive. The previous Objections have been withdrawn. However, there are new objections as shown above.
35 U.S.C. §112(b)
Applicant's arguments filed 4/14/2026, page 18, regarding the previous 112(b) Rejection of claim 25 have been fully considered and are persuasive. The previous 112(b) rejections have been withdrawn.
35 U.S.C. §103
Applicant's arguments filed 4/14/2026, pages 16-20, regarding the previous 103 Rejections of claims 12, 14-22, 24, 26, 28, and 30 have been fully considered but are 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.
Conclusion
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