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 .
Status of Claims and Rejections
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
Claims 1, 3, 9, and 13 have been amended.
Claims 14-21 were previously cancelled.
Claims 1-13 are pending.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1-13 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 1 recites “measuring…tissue between a liver and a transducer of the ultrasound scanner; scanning…a region of interest in the liver; determining an estimated fat fraction of the liver by a first machine-learned model configured to receive the measurement of the tissue and information from the scanning….”
The combined use of “measurement” and “information” renders claim 1 unclear such that a person having ordinary skill in the art could not determine the scope of the claim with reasonable certainty. The scope of the terms “measurement” and “information” are not clarified in the claims and are used so broadly in the specification that it is not clear what limits are made upon the ultimate inputs of the machine-learned model. While breadth is not to be equated with indefiniteness (MPEP 2173.04), each of these terms overlaps with one another and can be construed very broadly. With respect to the intervening tissue, “[t]he measurement is of the types of tissue, number of layers, thickness of each layer, acoustic characteristics of each layer, and/or other information layer-by-layer.” ([0040]). With respect to the ROI, “[t]he information from the ROI is the information from the liver used to calculate UDFF or other quantification (e.g., elasticity). The information from the intervening tissue is information that accounts for losses and/or wave distortions caused by the tissue, which may influence the quantification in the ROI.” ([0048]). Applicant’s disclosure even equates the meanings of measurement and information. “[T]he image processor 340 is configured by the machine-learned model 355 to output the quantification in response to input of the information from the ROI and information from intervening tissue.” ([0070]).
To illustrate the indefiniteness, the measurement could be a measured distance from the transducer to the liver, a particular parameter of a single layer of muscle and/or fat calculated by a formula using ultrasound data relating to the intervening tissue, and/or other information. The information of the ROI could be any information from a scan by an ultrasound scanner.
In addition to the above, the term “tissue” could refer to all tissue between the liver and the transducer or could refer to a single layer of muscle or fat between the liver and the transducer.
Combining these two very broad and overlapping meanings with the breadth of tissue, a person having ordinary skill in the art could not determine the scope of the claim with reasonable certainty.
For the purposes of a compact prosecution, Examiner is interpreting the relevant portions of claim 1 as follows:
…measuring, by the ultrasound scanner, a measurement of intervening tissue between a liver and a transducer of the ultrasound scanner, the measurement being derived from ultrasound data and including at least one of a thickness, a backscatter coefficient, or an attenuation of the intervening tissue;
scanning, by the ultrasound scanner, a region of interest in the liver to acquire quantitative ultrasound information based on scattering of the liver within the region of interest;
determining an estimated fat fraction of the liver by a first machine-learned model configured to receive the measurement of the intervening tissue and the quantitative ultrasound information, compensate for ultrasound signal losses and wave distortions caused by acoustic properties of the intervening tissue based on the measurement, and output the estimated fat fraction of the liver based on the measurement and the quantitative ultrasound information….
Claims 2-13 depend directly or indirectly from claim 1 and, as such, are also indefinite.
Further clarification is needed to overcome the rejection.
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
Claims 1-13 are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the enablement requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to enable one having ordinary skill in the art to which it pertains, or with which it is most nearly connected, to make and/or use the invention.
For the similar reasons as discussed above, claims 1-13 are non-enabling. Based on the indefiniteness caused by the meanings of “measurement” and “information” and “tissue”, one having ordinary skill in the art would not know how to make and/or use the invention.
RESPONSE TO APPLICANT’S ARGUMENTS
Applicant’s arguments filed with the response dated June 15, 2026 have been considered but are not persuasive. Applicant acknowledges that measurement and information are “broad terms,” but then alleges that the terms are clear because the terms are tied to subsequent limitations. “By tying the measurement to a specific compensation function, and tying the output to both the measurement and the scanning information, the claim architecture is structurally sound.” (p.6 of Remarks).
Examiner disagrees that the terms are sufficiently clear. Neither term is positively recited in the claims. Each term is introduced as something from a previous step (“determining an estimated fat fraction of the liver by a first machine-learned model configured to receive the measurement of the tissue and information from the scanning…”). Second, the clarity allegedly added by the amendments only indirectly affects the meaning of the terms. The claims do not positively define or clarify the meanings of measurement and information. For example, the term “information” alone from an ultrasound scanner does not require that the information be quantitative ultrasound information.
Accordingly, claims 1-13 are indefinite and not enabled.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-13 rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The claims recite: determining an estimated fat fraction of the liver using a first machine-learned model configured to receive the measurement of the tissue and information from the scanning, compensate for ultrasound signal losses and wave distortions caused by acoustic properties of the tissue based on the measurement, and output the estimated fat fraction of the liver based on the measurement and the information from the scanning.
The determining operation, as drafted and under its broadest reasonable interpretation, recites a mental process and/or a mathematical concept. It recites a mental process because the determining operation can be performed in the human mind. (see MPEP § 2106.04(a)(2)(III)). Examples of mental processes include “observations, evaluations, judgments, and opinions.” (Id). In this case, evaluating ultrasound image data (e.g., in which the evaluation is based on measurements of the intervening tissue and information from scanning, such as image data) can be a human cognitive action that has been performed for decades. The fact that a “machine-learned model” is used does not negate the judicial exception. (see MPEP § 2106.04(a)(2)(III) (explaining that claims requiring a computer still may recite a mental process) (see also the recently decided Recentive Analytics, Inc. v. Fox Corp., No. 2023-2437, Federal Circuit, 18 April 2025: “[C]laims that do no more than apply established methods of machine learning to a new data environment” are not patent eligible).
The determining operation also recites a mathematical concept because determining an estimated fat fraction of the liver includes using mathematical calculations that are based on attenuation and backscatter from ultrasound data. (see, e.g., the functions that used to calculate UDFF in [0056] and [0057] of LABYED ‘323 discussed below). Indeed, claim 1 explicitly recites compensating for ultrasound signal losses and wave distortions caused by acoustic properties of the tissue. Again, the fact that a “machine-learned model” is used does not negate the judicial exception. (see MPEP § 2106.04(a)(2)(III) (explaining that claims requiring a computer still may recite a mental process). See CyberSource Corp. v. Retail Decisions, Inc.,, 654 F.3d at 1376, 99 USPQ2d at 1373 (“[C]omputational methods which can be performed entirely in the human mind are the types of methods that embody the ‘basic tools of scientific and technological work”‘).
The judicial exception is not integrated into a practical application. In this case, the additional elements include measuring, by the ultrasound scanner, tissue between a liver and a transducer of the ultrasound scanner and scanning, by the ultrasound scanner, a region of interest in the liver. However, the measuring and scanning operations are pre-solution activity that must occur in order to determine the estimated fat fraction (i.e., to perform the judicial exception). Moreover, each of these generically recites a step of quantitative ultrasound (i.e., measuring a property of the tissue). The ultrasound scanner is also generically recited without any meaningful limitations to its scope. Likewise, the relevant data (i.e., measurement and information) that determine the estimated fat fraction are recited without any specificity. As such, these additional elements only generally link the use of the judicial exception to the technological environment of quantitative ultrasound.
The additional elements also include displaying the ultrasound derived fat fraction. However, this is merely post-solution activity (i.e., displaying a result of the judicial exception). Accordingly, each of the additional elements adds insignificant extra-solution activity to the judication exception. (see MPEP § 2106.05(g)).
Lastly, the claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above, the additional elements generically recite collecting data and then, after performing the judicial exception, displaying a result. The additional elements do not meaningfully limit the abstract idea. Accordingly, the claims essentially recite a claim to “collecting information, analyzing it, and displaying certain results of the collection and analysis,” where the data analysis steps are recited at a high level of generality such that they could practically be performed in the human mind, Electric Power Group v. Alstom, S.A., 830 F.3d 1350, 1353-54, 119 USPQ2d 1739, 1741-42 (Fed. Cir. 2016).
Dependent claims 2-13 do not render the subject matter patent eligible. While claim 2 does specify that at least one measurement includes a thickness, the claim does not specify if the thickness is of a particular tissue. It is insignificant pre-solution activity because measuring a thickness of tissue is commonly done in many ultrasound protocols where the propagation path goes through tissue.
With respect to claim 3, measuring a backscatter coefficient and/or attenuation of the tissue does not integrate the judicial exception into a practical application or meaningfully limit the claim. Backscatter coefficient and/or attenuation is measured in many quantitative ultrasound processes (see, e.g., FERRAIOLI and KRISHNAN discussed with respect to the Section 103 rejections). It is insignificant pre-solution activity. As to the latter recitation of claim 3 (and in light of the Section 112 rejection), knowing the location of measurements is necessary when calculating parameters. (see, e.g., discussion regarding depth in FERRAIOLI and KRISHNAN).
Claim 4 recites measuring a location along the propagation path and a graphical user interface object. Both limitations (i.e., measuring tissue that the propagation path must go through) and positioning graphical objects over the B-mode image are well-known and performed in many ultrasound protocols. (see, e.g., Section 103 rejection based on ALI below). Furthermore, as discussed above, the step of measuring tissue is insignificant pre-solution activity. Placing graphical objects on landmarks does not meaningfully limit the claims.
Claims 5 and 6 recite limitations to the step of measuring. However, both are generically recited in the context of ultrasound imaging. For example, claim 5 recites measuring an acoustic property of the tissue. Ultrasound is fundamentally based upon acoustic properties. With respect to claim 6, identifying different tissue layers of the tissue and wherein the measurement is derived from the different tissue layers is known. (see, e.g., rejection of claim 6).
Claim 7 does not meaningfully limit the claim as all measurements along the propagation path of the tissue would likely be considered for the estimated fat fraction determination. Claim 7 recites that “characteristic” of each layer is input but, as described in Applicant’s specification, the term has a very broad meaning. As such, it does not meaningfully limit the judicial exception.
Claim 8 recites that the machine-learned model is configured by training to account for losses and/or wave distortions caused by the tissue. However, as best understood by the examiner, estimated fat fraction measurements inherently account for losses and/or wave distortions caused by tissue. As such, it would be necessary for the machine-learned model to consider them. As such, it does not meaningfully limit the judicial exception.
Claim 9 recites that a location of a liver capsule is automatically detected and graphical objects are automatically placed on the image. However, automatic detection is known (see discussion regarding Section 103 rejection of claim 9). Moreover, automating processes that have historically been performed by humans does not meaningfully limit the abstract idea.
Claim 10 recites examining a field of view by a second machine-learned model and outputting guidance to position the transducer to image the liver based on output of the second machine-learned model. Claim 9 recites a judicial exception (i.e., mental process) of judging the field of view to determine whether a position of a transducer should be changed.
Claim 11 recites that examining includes examining for shadows and/or vessels, wherein the guidance reduces the shadows and/or vessels in the field of view. However, as best understood by the examiner, examining for shadows and/or vessels is what must be done in order to acquire sufficient data. As such, it does not meaningfully limit the judicial exception.
Claim 12 recites that the examining comprises scoring the field of view for automated placement of the region of interest. However, scoring a field of view recites another mental process (e.g., mathematical concept) and does not integrate the judicial exceptions into a practical application.
Claim 13 recites determining the estimated fat fraction as a field of estimated fat fraction values distributed in a region of interest in the liver. This recites a mental process (e.g., mathematical concept) and does not integrate the judicial exceptions into a practical application. Furthermore, displaying an ultrasound image with the region of interest coded by the estimated fat fraction values is insignificant extra-solution activity.
Accordingly, none of the claims are patent ineligible.
RESPONSE TO APPLICANT’S ARGUMENTS
With respect to the Section 101 rejection, Applicant argues that the claimed subject matter is integrated into a practical application because it provides a specific improvement to a technology. (see page 7 of Remarks). Applicant argues that “Applicants' specification identifies a specific technological problem in the field of ultrasound imaging: quantitative evaluation of tissue properties is influenced by the type, thickness, and alignment of intervening tissue layers. Attenuation in these intervening layers causes signal losses and miscalculations, resulting in less accurate quantification of the liver.”
Examiner disagrees. Applicant’s argument relies upon limitations that are not currently found in the claim 1 (see, e.g., type, thickness, and alignment of intervening tissue layers). However, “…if the specification sets forth an improvement in technology, the claim must be evaluated to ensure that the claim itself reflects the disclosed improvement.” (MPEP 2106.04(d)(1)). Again, the terms “measurement,” “information,” and “tissue” are recited broadly such that the claim encompasses more than the disclosed improvement and there are no limitations as to how compensation is implemented. Claim 1 is merely using the words “apply it” (or an equivalent) in combination with a machine-learned model. (MPEP 2106.04(d), I).
Applicant also argues that “utilizing a trained machine-learned model cannot practically be performed in the human mind….” First, this argument suggests that all claims reciting a machine-learned model would be patent eligible. However, this is not the case. (see Recentive Analytics, Inc. v. Fox Corp., No. 2023-2437, Federal Circuit, 18 April 2025: “[C]laims that do no more than apply established methods of machine learning to a new data environment” are not patent eligible). Second, courts do not “distinguish between claims that recite mental processes performed by humans and claims that recite mental processes performed on a computer.” (MPEP 2106.04(a)(2), III; see also part C: “Claims can recite a mental process even if they are claimed as being performed on a computer.”).
Lastly, the claims are recited at a high level of generality without specificity, which is a hallmark of ineligible subject matter. (MPEP 2106.04(a)(2), III, C; see, e.g., “Berkheimer v. HP, Inc., 881 F.3d 1360, 125 USPQ2d 1649 (Fed. Cir. 2018), in which the patentee claimed methods for parsing and evaluating data using a computer processing system. The Federal Circuit determined that these claims were directed to mental processes of parsing and comparing data, because the steps were recited at a high level of generality and merely used computers as a tool to perform the processes.”; see also Electric Power Group v. Alstom, S.A., 830 F.3d 1350, 1353-54, 119 USPQ2d 1739, 1741-42 (Fed. Cir. 2016). (MPEP 2106.04(d)). For example, “measurement” can be interpreted broadly to include physical distances or calculated parameters, such as AC and BSC. Likewise, information could be any ultrasound data or parameters calculated from the ultrasound data, such as AC, BSC, or other parameters. As another example, the tissue that is measured could be all tissue between the liver and the transducer or just one layer of tissue (e.g., muscle, fat, skin).
Accordingly, the claims are not patent ineligible.
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 1-5 and 8 are rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent Appl. Publ. No. 2018/0289323 A1 to Labyed (hereinafter “LABYED ‘323”) and Ferraioli, Giovanna, et al. “Liver fat quantification with ultrasound: depth dependence of attenuation coefficient.” Journal of Ultrasound in Medicine 42.10 (2023): 2247-2255 (hereinafter “FERRAIOLI”) and/or Krishnan KB, Nagaraj N, Singhal N, Thapar S, Yadav K. A two-parameter model for ultrasonic tissue characterization with harmonic imaging. arXiv preprint arXiv:1712.03495. 2017 Dec 10. (hereinafter “KRISHNAN”) and in further view of U.S. Patent Appl. Publ. No. 2023/0100912 A1 to Amar (hereinafter “AMAR”).
With respect to claim 1 (and in light of the Section 112(a) and (b) rejections), LABYED ‘323 teaches a method for ultrasound imaging with an ultrasound scanner. LABYED ‘323 teaches embodiments for ultrasound imaging that measure a tissue property, such as liver fat fraction. ([0002]). LABYED ‘323 teaches:
scanning, by the ultrasound scanner, a region of interest in the liver. “A medical diagnostic ultrasound scanner performs the measurements by acoustically generating the waves and measuring the responses.” ([0016]). “[T]he tissue property is indicated in a region of interest (sub-part of the field of view) or over the entire field of view.” ([0062]).
determining an estimated fat fraction of the liver by a first machine-learned model configured to receive…information from the scanning and output the estimated fat fraction. LABYED ‘323 teaches a “machine-learnt classifier” that estimates a tissue property. (see, e.g., [0049]). “Any machine learning and resulting machine-learnt classifier may be used.” ([0049]; see also [0050]-[0053]). The tissue property may be an estimated fat fraction. “Any tissue property may be estimated. For example, the fat fraction of tissue is estimated.” ([0047]). Fat fraction of tissue includes ultrasound-derived fat fraction (UDFF). (see, e.g., [0055]). The classifier receives input values of the ultrasound parameters and outputs the estimated fat fraction. “The percentage of fat is used as the ground truth so that the machine learning learns to classify the percentage of fat from input values for the ultrasound parameters.” ([0050]; see also [0051], [0052], and [0078]). Notably, the machine-learning model is trained to learn “what weighted combination (e.g., convolution using learnt kernel) of input values indicates the output.” ([0051]).
displaying the estimated fat fraction. “In act 38, the ultrasound scanner or a display device displays the estimated tissue parameter. For example, an image of the fat fraction is generated.” ([0057]).
However, LABYED ‘323 does not explicitly teach measuring, by the ultrasound scanner, tissue between a liver and a transducer of the ultrasound scanner and that the first machine-learned model is configured to receive the measurement of the tissue in addition to the information from the scanning. Nonetheless, LABYED ‘323 does teach that the machine-learning model (i.e., classifier) may be trained using other measurements or information. (see, e.g., [0046] and [0050]). More specifically, LABYED ‘323 teaches that any information about the patient may be used for estimating the tissue property. ([0046]). “For example, clinical information for the patient is used. The clinical information may be the medical history, age, body-mass index, sex, fasting or not, blood pressure, diabetic or not, and/or a blood biomarker measure…Any information about the patient may be included.” ([0046]). Importantly, the machine-learning model in LABYED ‘323 is trained to learn “what weighted combination (e.g., convolution using learnt kernel) of input values indicates the output.” ([0051]).
In the same field of endeavor, FERRAIOLI conducted a study that estimated the influence of various liver depths on the attenuation coefficient of various QUS vendors. (Abstract). FERRAIOLI teaches that the attenuation coefficient, one of the parameters use to determine an estimated fat fraction, is depth dependent. (Abstract). FERRAIOLI studied three different systems by Canon, Philips, and Siemens. (p.2248, right column). For each system, FERRAIOLI measured the skin-to-liver capsule distance (SCD), which is a measurement that includes the tissue between the liver and the ultrasound transducer. FERRAIOLI found that “[t]he results of this study show that the AC values depend on the depth of the measurement and that there is a progressive decrease of the values that is directly related to the depth. This finding is of utmost relevance because thresholds for detecting and grading liver steatosis might vary depending on the ROI’s depth for AC measurements.” (p.2253, Discussion).
FERRAIOLI also taught that the ROI should be close to the “elevational focus” of the transducer. “ROI close to elevational focus should likely be the default setting because AC is overestimated at depths lower than elevational focus and underestimated at depths higher than elevational focus.” (p.2254, right column). FERRAIOLI noted that in one earlier study of a Canon system the AC measurements with the highest repeatability were acquired by a transducer with an elevational focus that coincided with a center of the ROI. This occurred because the recommended placement of the ROI below the liver capsule (i.e., 2 cm below capsule) plus the mean thickness of the subcutaneous tissue was about equal to the elevational focus. In other words, for the ROI to be close to the elevational focus as recommended by FERRAIOLI, one should consider the depth below the liver capsule and the thickness of the subcutaneous tissue.
FERRAIOLI also found that the skin-to-liver capsule distance (SCD) affected the AC values for two systems that had about the same SCD, including Siemens, and that “[b]ody mass index” and “waist circumference” were strongly correlated to the skin-to-liver capsule distance. (p.2252, right column). Both BMI and waist circumference are associated with a thicker subcutaneous layer and non-alcoholic fatty liver.
Accordingly, FERRAIOLI teaches that AC measurements depend upon the depth and size of ROI and that a thickness of the subcutaneous tissue correlates to BMI and waist circumference and can affect AC measurements.
In the same field of endeavor, KRISHNAN teaches a method that has been “approximated and generalized to estimate AC and BSC for tissue layer underlying a more attenuative subcutaneous layer.” (Abstract). KRISHNAN proposes a model that is based, in part, on the attenuation coefficient and the backscatter coefficient of the subcutaneous layer. “[T]he attenuation of the ultrasound beam through the subcutaneous layer of thickness 𝛿 in the Sample tissue where ∝𝑓 (in units of Neper/cm) is the attenuation coefficient of the layer. The backscatter coefficient of the subcutaneous layer is assumed to be 𝐵𝑆𝐶𝑆, the same as that of the Sample tissue.” (p.7, 3.1 Extension of Basic Model to Practice). KRISHNAN notes that errors may occur in the model at larger thicknesses of the subcutaneous layer. “However, errors could accrue at larger distal depths 𝑧2 when 𝛼𝑆>𝛼𝑅 with increase in the thickness of the subcutaneous layer 𝛿.” (p.8, prior to 3.2). More specifically, the attenuation of the sample tissue is greater than the attenuation of the reference tissue due to the greater depth.
KRISHNAN also teaches that the thickness can be determined by viewing the image produced by the ultrasound scanner. (p.16, bottom). However, KRISHNAN notes that the identification of the subcutaneous layer can be “automated and improved by incorporating learning-based methods.” (p.22).
It would have been obvious to one having ordinary skill in the art to modify the ML model of LABYED ‘323 to account for the tissue between a liver and a transducer of the ultrasound scanner (i.e., propagation loss correction) such that the ML model received a measurement of the tissue in addition to the information from the scanning. LABYED ‘323 teaches that other information or measurements may be considered by the machine-learning model and also teaches that the attenuation coefficient, one of the parameters of estimated fat fraction, is a function of depth. ([0025]). FERRAIOLI and KRISHNAN confirm that the subcutaneous layer can affect AC measurements and KRISHNAN teaches that the subcutaneous layer can affect the BSC measurements. An estimated fat fraction, such as UDFF, can be based upon AC and BSC measurements. Accordingly, it would have been obvious to one having ordinary skill in the art to use a trained machine-learning model that outputs an estimated fat fraction in response to receiving an input measurement of the subcutaneous layer (e.g., at least subcutaneous layer thickness, AC measurement, and/or BSC measurement) and information from scanning the liver. One would have been motivated to use such a machine-learned model to correct or otherwise account for the affect that the intervening tissue has on liver quantification in order to obtain a more accurate estimated fat fraction. There would have been a reasonable expectation of success as LABYED ‘323 teaches that machine-learned models can be trained using various information.
NOTE: FERRAIOLI was published in 2023 and appears to include information that was provided by Siemens. Even if an inventor provided the information, the above analysis is still applicable because FERRAIOLI studied two other systems in addition to the Siemens system. Examiner also notes it is known that the output of a third system, FibroScan by Echosens, is improved when an adjustment is made based on the skin-capsular distance (SCD). (Kimura, Syunichiro, et al. “Effect of skin capsular distance on controlled attenuation parameter for diagnosing liver steatosis in patients with nonalcoholic fatty liver disease.” Scientific reports 11.1 (2021): See, e.g., Abstract: “Adjustment of the CAP using the SCD improves the diagnostic performance of the CAP in NAFLD.”). Regardless, KRISHNAN’s description of the effect that the subcutaneous layer has on measurements is still applicable and further supports using a measurement of the intervening tissue as an input for the machine-learning model.
However, LABYED ‘323 does not explicitly teach that the first machine-learned model is also configured to compensate for ultrasound signal losses and wave distortions caused by acoustic properties of the tissue based on the measurement. Nonetheless, KRISHNAN teaches compensating for the excess attenuation that is caused by the thickness of the intervening subcutaneous layer. “Equation (9) may be rewritten by scaling the 𝐼𝑆 term at 𝑧2 by compensating for the excess attenuation ∝𝑓−∝𝑅 with respect to the attenuation of the Reference tissue through 𝛿….” (p.7, see also Equation 10). FERRAIOLI further teaches that a thickness of the intervening tissue can affect measurements for estimating fat fraction. (p.2252, right column).
AMAR teaches various approaches for delivering ultrasound energy to a target region during a diagnostic procedure. (Abstract). AMAR is concerned with error caused by intervening tissue. (see, e.g., [0007] and [0009]). To address these concerns, AMAR teaches that ultrasound energy can be adjusted by a machine-learning model that has been trained “to compensate for expected beam aberration resulting for an intervening tissue….” (Abstract). “Approaches described herein advantageously mitigate or eliminate measurement errors having a common source but potentially affecting different transducer elements differently, so that different measurements may exhibit different errors in the measured parameter(s), e.g., phase, amplitude and/or time delay.” ([0012]).
Notably, the inputs to the machine-learned model include tissue characteristics of the intervening tissue. “The adjustment mechanism is configured (i) to receive numeric quantities (e.g., tissue density and/or thickness) characterizing tissue intervening between the active transducer elements and the target region….” ([0018]; see also [0072]: “The numeric quantities may include anatomic features (e.g., thickness and/or density) and/or tissue characteristics associated with the target and/or intervening tissue and/or a geometric parameter (e.g., beam-skull angle) associated with each of the transducer element.”). The adjustment mechanism compensates for ultrasound signal losses and wave distortions caused by acoustic properties of the tissue based on the measurement. “In one implementation, the generated parameter value(s) specifies a correction to an amplitude, a phase, a frequency, a duty cycle, a sonication pattern and/or a time delay.” ([0018]). While AMAR specifically concerns therapeutic ultrasound, AMAR teaches that its invention could be applied to ultrasound imaging. ([0011]).
It would have been obvious to one having ordinary skill in the art at the time of filing to modify the LABYED ‘323 system such that the first machine-learned model compensates for ultrasound signal losses and wave distortions caused by acoustic properties of the tissue based on the measurement, as taught in AMAR. AMAR specifically teaches that measurements of an intervening tissue (e.g., thickness of intervening tissue) can be input into a machine-learning model to compensate for ultrasound signal losses and wave distortions caused by the intervening tissue. Modifying the first machine-learned model to receive a measurement of the intervening tissue would be applying a known technique to a known method that is ready for improvement to yield predictable results. Moreover, one of ordinary skill in the art would have been motivated to make this modification because it is known that the intervening tissue causes error and that the error can be corrected by using a measurement as an input. There would have been a reasonable expectation of success as AMAR teaches that a machine-learned model can receive and use a measurement of the intervening tissues as an input.
With respect to claim 2, LABYED ‘323 does not explicitly teach that measuring comprises measuring a thickness as the measurement. However, as discussed above with respect to claim 1, KRISHNAN specifically teaches extracting the thickness of the subcutaneous fat layer based on images from the ultrasound scanner. (See bottom of p.16 and Figure 7). KRISHNAN teaches determining this thickness in order to account for the attenuation caused by the subcutaneous fat. (see, e.g., Abstract). Moreover, AMAR specifically teaches that measurements of an intervening tissue (e.g., thickness of intervening tissue) can be input into a machine-learning model to compensate for ultrasound signal losses and wave distortions caused by the intervening tissue.
It would have been obvious to one having ordinary skill in the art at the time of filing to modify the LABYED ‘323 system such that the measurement included a thickness of the intervening tissue, as taught in AMAR. AMAR specifically teaches that measurements of an intervening tissue (e.g., thickness of intervening tissue) can be input into a machine-learning model to compensate for ultrasound signal losses and wave distortions caused by the intervening tissue. Modifying the first machine-learned model to receive a measurement of the intervening tissue would be applying a known technique to a known method that is ready for improvement to yield predictable results. Moreover, one of ordinary skill in the art would have been motivated to make this modification because it is known that the intervening tissue causes error and that the error can be corrected by using a measurement as an input. There would have been a reasonable expectation of success as AMAR teaches that a machine-learned model can receive and use a measurement of the intervening tissues as an input.
With respect to claim 3, LABYED ‘323 does not explicitly teach that measuring comprises measuring a backscatter coefficient and/or attenuation of the tissue as the measurement or that the determining comprises the machined-learned model receiving the measurement of the tissue as the backscatter coefficient and/or attenuation of the tissue. However, LABYED ‘323 explicitly teaches that the estimated fat fraction of the liver may be calculated using the attenuation coefficient and the backscatter coefficient. (see, e.g., [0056]).
In the same field of endeavor, KRISHNAN teaches configuring the model to consider the AC and the BSC of the subcutaneous fat layer in order to more accurately predict disease progression. (Abstract; see also “3.1 Extension of Basic Model to Practice” on p.7 in which the model is adapted based on the attenuation coefficient and the backscatter coefficient of the subcutaneous layer.)
It would have been obvious to one having ordinary skill in the art to include a measurement of a backscatter coefficient and/or attenuation coefficient of the intervening tissue and provide the backscatter coefficient and/or attenuation coefficient of the tissue to the machine-learned model. LABYED ‘323 teaches that other information or measurements may be considered. KRISHNAN confirm that the subcutaneous layer can affect the determination of disease progression of the liver and that at least one of the AC or the BSC can affect that determination. One would have been motivated to configure the machine-learning model to receive the backscatter coefficient and/or attenuation of the tissue in order to output a more accurate estimated fat fraction. There would have been a reasonable expectation of success as LABYED ‘323 teaches that machine-learned models can be trained using various information.
With respect to claim 4, LABYED ‘323 in view of FERRAIOLI and KRISHNAN teach that the measuring comprises measuring the tissue between a liver capsule of the liver and the transducer. As discussed above in the rejection of claim 1, KRISHNAN teaches measuring the thickness of the subcutaneous fat layer. FERRAIOLI also teaches that the liver capsule is indicated by an indicator on a display. Figure 1C on p.2249 of FERRAIOLI shows a horizontal line at the liver capsule. (Horizontal line is even with the cross-hairs to the left of the image.)
NOTE: While Figure 1C is shown in FERRAIOLI and possibly provided by the inventor, the image is consistent with other images that were publicly available before 2023. For example, the previously issued Office Action relied upon Figure 3a of GAO, which shows an indicator for the liver capsule. GAO used Siemens’s Sequoia system, which is the same system used in FERRAIOLI.
It would have been obvious to one having ordinary skill in the art to indicate the liver capsule on a display as illustrated in FERRAIOLI and/or GAO. One would have been motivated to include the indicator to guide and assure the technician that the ROI is properly positioned away from the liver capsule while acquiring measurements. There would have been a reasonable expectation of success as FERRAIOLI and/or GAO teach that user displays can include such indicators.
With respect to claim 5, LABYED ‘323 does not explicitly teach that the measuring comprises measuring an acoustic property based on a type of the tissue as the measurement. However, as explained above with respect to claim 1 and claim 3, KRISHNAN teaches configuring the model to consider the AC and the BSC of the subcutaneous fat layer in order to more accurately predict disease progression. (Abstract; see also “3.1 Extension of Basic Model to Practice” on p.7 in which the model is adapted based on the attenuation coefficient and the backscatter coefficient of the subcutaneous layer.)
It would have been obvious to one having ordinary skill in the art to include measuring at least one acoustic property of the subcutaneous fat layer. KRISHNAN confirms that the subcutaneous layer can affect the determination of disease progression of the liver and that the thickness and at least one of the AC or the BSC can affect that determination. One would have been motivated to train the machine-learning model to account for the thickness and/or an acoustic property of the subcutaneous fat layer when outputting an estimated fat fraction because considering how the intervening tissue affects the ultrasound signal would provide a more accurate estimated fat fraction as taught by KRISHNAN.
With respect to claim 8, the combination of LABYED ‘323, FERRAIOLI, KRISHNAN, and AMAR teach that the machine-learned model is configured by training to account for losses and/or wave distortions caused by the tissue. As explained above with respect to claim 1 and claim 3, KRISHNAN teaches configuring the model to consider the AC and the BSC of the subcutaneous fat layer in order to more accurately predict disease progression. AMAR specifically teaches that measurements of an intervening tissue can be input into a machine-learning model to compensate for ultrasound signal losses and wave distortions caused by the intervening tissue. Accordingly, a machine-learned model trained to account for the thickness of the subcutaneous fat layer and AC and/or BSC measurements when outputting an estimated fat fraction would necessarily account for the losses and/or wave distortions caused by the tissue.
Claims 6 and 7 are rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent Appl. Publ. No. 2018/0289323 A1 to Labyed (hereinafter “LABYED ‘323”) and Ferraioli, Giovanna, et al. “Liver fat quantification with ultrasound: depth dependence of attenuation coefficient.” Journal of Ultrasound in Medicine 42.10 (2023): 2247-2255 (hereinafter “FERRAIOLI”) and/or Krishnan KB, Nagaraj N, Singhal N, Thapar S, Yadav K. A two-parameter model for ultrasonic tissue characterization with harmonic imaging. arXiv preprint arXiv:1712.03495. 2017 Dec 10. (hereinafter “KRISHNAN”) and in further view of U.S. Patent Appl. Publ. No. 2023/0100912 A1 to Amar (hereinafter “AMAR”) as applied to claim 1 above, and further in view of Wear et al. “US backscatter for liver fat quantification: an AIUM-RSNA QIBA pulse-echo quantitative ultrasound initiative.” Radiology 305.3 (2022): 526-537 (hereinafter “WEAR”).
With respect to claim 6, KRISHNAN teaches that the measuring comprises identifying different tissue layers of the tissue because identifying the subcutaneous fat layer necessarily distinguishes it from different tissue layers. None of LABYED ‘323, FERRAIOLI, or KRISHNAN explicitly teach that the measurement is derived from the different tissue layers. However, AMAR teaches that inputs may include the thickness and/or density of different layers. (see, e.g., [0046]).
In the same field of endeavor, WEAR is a review article that “explains the science and clinical evidence underlying backscatter for liver fat assessment. Recommendations for data collection are discussed, with the aim of minimizing potential confounding effects associated with technical and biologic variables.” (emphasis added) (Abstract). Notably, WEAR concerns measuring a backscatter coefficient (BSC), which is one of the parameters on which estimated fat fraction calculations are based. “Calculation of BSC requires compensation for the total attenuation of US by all intervening tissues between the body surface and the deepest point in the ROI in the liver.” (emphasis added) (p.534, right column, first paragraph of Section entitled “Compensation for Attenuation…”). One approach includes compensating for each individual tissue of these intervening tissues. “This approach relies on identification and measurement of tissue layers (i.e., skin, muscle, fat, and liver) in the propagation path to the ROI.” (emphasis added) (Id., following paragraph). For this approach, “[a]n AC value at each BSC measurement frequency is assigned to each tissue using representative values in the literature.” (Id).
WEAR teaches that “[c]alculation of BSC requires compensation for the total attenuation of US by all intervening tissues between the body surface and the deepest point in the ROI in the liver.” (p.534, right column, first paragraph of Section entitled “Compensation for Attenuation…”). One approach for compensating “relies on identification and measurement of tissue layers (i.e., skin, muscle, fat, and liver) in the propagation path to the ROI.” (emphasis added) (Id., following paragraph). KRISHNAN already teaches that one should consider the thickness, AC, and BSC of the subcutaneous fat layer.
It would have been obvious to one having ordinary skill in the art to identify different tissue layers of the tissue when measuring the intervening tissue and for the measurement to be derived from the different tissue layers. KRISHNAN already teaches distinguishing the subcutaneous fat layer and identifying a thickness and acoustic properties of the fat layer. AMAR and WEAR teach that one should consider each of the intervening tissues (e.g. skin, muscle, fat). One having ordinary skill in the art would have been motivated to train the machine-learning model to account for the thickness and/or an acoustic property of each of the intervening tissue layers when outputting an estimated fat fraction because considering how the intervening tissues affect the ultrasound signal would provide a more accurate estimated fat fraction as taught by KRISHNAN, WEAR, and AMAR.
With respect to claim 7 (depending from claim 6), WEAR and KRISHNAN teach wherein a characteristic of each of the different tissue layers is input to the first machine-learned model as the measurement.
As discussed above with respect to claim 6, AMAR teaches that inputs may include the thickness and/or density of different layers. (see, e.g., [0046]). Furthermore, WEAR teaches that “[c]alculation of BSC requires compensation for the total attenuation of US by all intervening tissues between the body surface and the deepest point in the ROI in the liver.” (p.534, right column, first paragraph of Section entitled “Compensation for Attenuation…”). One approach for compensating “relies on identification and measurement of tissue layers (i.e., skin, muscle, fat, and liver) in the propagation path to the ROI.” (emphasis added) (Id., following paragraph). KRISHNAN already teaches that one should consider the thickness, AC, and BSC of the subcutaneous fat layer.
It would have been obvious to one having ordinary skill in the art to identify different tissue layers of the tissue. It would have also been obvious to input a characteristic of each of the different tissue layers into the machine-learned model. KRISHNAN already teaches distinguishing the subcutaneous fat layer and identifying a thickness and acoustic properties of the fat layer. AMAR and WEAR teach that one should consider each of the intervening tissues (e.g. skin, muscle, fat). One having ordinary skill in the art would have been motivated to train the machine-learning model to account for the thickness and/or an acoustic property of each of the intervening tissue layers when outputting an estimated fat fraction because considering how the intervening tissues affect the ultrasound signal would provide a more accurate estimated fat fraction as taught by KRISHNAN, WEAR, and AMAR.
Claim 9 is rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent Appl. Publ. No. 2018/0289323 A1 to Labyed (hereinafter “LABYED ‘323”) and Ferraioli, Giovanna, et al. “Liver fat quantification with ultrasound: depth dependence of attenuation coefficient.” Journal of Ultrasound in Medicine 42.10 (2023): 2247-2255 (hereinafter “FERRAIOLI”) and/or Krishnan KB, Nagaraj N, Singhal N, Thapar S, Yadav K. A two-parameter model for ultrasonic tissue characterization with harmonic imaging. arXiv preprint arXiv:1712.03495. 2017 Dec 10. (hereinafter “KRISHNAN”) and in further view of U.S. Patent Appl. Publ. No. 2023/0100912 A1 to Amar (hereinafter “AMAR”) as applied to claim 1 above, and further in view of Gao et al. “Reliability of performing ultrasound derived SWE and fat fraction in adult livers.” Clinical imaging 80 (2021): 424-429 (hereinafter “GAO”) and U.S. Patent Appl. Publ. No. 2024/0398382 A1 (hereinafter “ALI”).
With respect to claim 9, none of LABYED ‘323, FERRAIOLI, KRISHNAN, and AMAR teach the claim limitations. However, in the same field of endeavor, GAO describes a study to test the reproducibility of performing conventional point shear wave elastography (pSWE), auto-pSWE, and ultrasound derived fat fraction (UDFF) in adult livers. (Abstract). “An Acuson Sequoia ultrasound scanner (Siemens Healthineers, Mountain View, CA, USA) equipped with a curvilinear probe (5C1, bandwidth 1.0-5.7 MHz) was used to acquire grayscale imaging and measure ultrasound SWE and fat fraction (UDFF) parameters of the liver.” (p.425, left column). As shown in Figures 3a-c below, GAO teaches placing a line from the transducer through the liver capsule (see vertical line extending through image), an indicator on the liver capsule
PNG
media_image1.png
257
713
media_image1.png
Greyscale
(overlaying text identifies this line as “liver capsule” in Figure 3a), and the region of interest in the liver along the line (ROI has curved trapezoid-like shape), and wherein determining the estimated fat fraction comprise determining the estimated fat fraction in the region of interest. “The region of interest (ROI, 3.0 cm × 3.0 cm, laterally by axially) for measuring ultrasound derived fat fraction (UDFF %) is placed in the liver.” (see caption of Figure 3a).
However, GAO does not explicitly teach that the line from the transducer, the indicator of on the liver capsule, or the region of interest in the liver along the line are automatically placed into position. Moreover, GAO does not explicitly teach automatically detecting a location of a liver capsule in an ultrasound image of the liver.
In the same field of endeavor, ALI teaches ultrasound imaging techniques for shear-wave elastography. (Abstract). ALI notes that shear-wave elastography can be used to measure or estimate stiffness of the liver, which is one sign of fatty liver disease. The techniques include automatically identifying an area for a region of interest within the segmented region, wherein the region of interest corresponds to a region in the liver for performing shear-wave elastography. (Abstract).
In ALI, “[t]he region of interest may be automatically located or location may be facilitated through an automatic process.” ([0023]). The ROI can be based on the location of the liver capsule. ([0023]). In one embodiment, an AI model is trained to automatically identify the liver capsule. ([0049]). “For example, a model may be trained to segment a liver capsule or liver, and then the trained model may be included in ultrasound system 100, for example, as part of signal processor 132 or associated memory.” ([0050]). The ROI can then be identified based on the location of the liver capsule. ([0086]). Figure 14 shows an indictor 1108 (horizontal line) over the liver capsule that is automatically positioned over the liver capsule and update during imaging. (see [0095], [0096] and also [0090]).
It would have been obvious to one having ordinary skill in the art to modify the system to automatically detect a location of a liver capsule in an ultrasound image of the liver and automatically place the relevant indicators (i.e., the line from the transducer, the indicator of on the liver capsule, and the region of interest in the liver along the line), as taught in ALI. One having ordinary skill in the art would be motivated to make this modification because automatic placement of the relevant indicators expedites the measuring process for the technician. There would have been a reasonable expectation of success because ALI teaches that a system can be capable of automatically detecting the liver capsule and ROI and positioning indicators with respect to them.
Claims 10-12 are rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent Appl. Publ. No. 2018/0289323 A1 to Labyed (hereinafter “LABYED ‘323”) and Ferraioli, Giovanna, et al. “Liver fat quantification with ultrasound: depth dependence of attenuation coefficient.” Journal of Ultrasound in Medicine 42.10 (2023): 2247-2255 (hereinafter “FERRAIOLI”) and/or Krishnan KB, Nagaraj N, Singhal N, Thapar S, Yadav K. A two-parameter model for ultrasonic tissue characterization with harmonic imaging. arXiv preprint arXiv:1712.03495. 2017 Dec 10. (hereinafter “KRISHNAN”) and in further view of U.S. Patent Appl. Publ. No. 2023/0100912 A1 to Amar (hereinafter “AMAR”) as applied to claim 1 above, and further in view of U.S. Patent Appl. Publ. No. 2021/0177373 A1 to Xie et al. (hereinafter referred to as “XIE”).
With respect to claim 10, none of LABYED ‘323, FERRAIOLI, and KRISHNAN explicitly teach examining a field of view by a second machine-learned model and outputting guidance to position the transducer to image the liver based on output of the second machine-learned model.
In the same field of endeavor, XIE teaches “ultrasound imaging systems and methods for ultrasonically inspecting biological tissue, such as liver and for automatically identifying and acquiring a view suitable for hepatic-renal echo-intensity ratio quantification, using one or more neural networks….” (Abstract). XIE notes that “[u]ltrasound imaging can be used to measure [a quantitative biomarker], but it may be subject to misdiagnoses or classification due to difficulties in achieving the proper frame for measurement purposes.” ([0002]).
As such, XIE teaches an intelligent liver scan mode 101 in which “the system may execute one or more sets of instructions for view matching, automated image capture, ROI identification, and echo-intensity ratio quantification.” ([0029]). XIE’s system examines a field of view. “During this process, the system may determine and/or output (e.g., for display to the user) a confidence metric, which is indicative of the live ultrasound image corresponding to a view suitable for H/R ratio quantification….” (Id). This examination may be performed by a machine-learned model. “[T]he system may be trained or otherwise configured to recognize whether the image corresponds to a suitable image view, also referred to as target image view. In some embodiments, the view matching sub-process may be performed or enhanced with one or more machine learning image classification models (block 115).” XIE also teaches outputting guidance to position the transducer to image the liver based on output of the second machine-learned model. “In some example, the system may guide the user in acquiring the appropriate view of the tissue. FIG. 4, panels a-c shows additional graphical displays that may be provided during the AI-assisted liver scan.” (emphasis added) ([0049]).
It would have been obvious to one having ordinary skill in the art to modify the system to utilize a machined-learned model for directing the user to capture ultrasound images. One having ordinary skill in the art would be motivated to use a machine-learned model to increase the likelihood that the operator will position the probe accurately and avoid misdiagnoses as taught in XIE. There would have been a reasonable expectation of success because XIE demonstrates that such systems can incorporate AI assistance.
With respect to claim 11 (depending from claim 10), none of LABYED ‘323, FERRAIOLI, KRISHNAN, and AMAR explicitly teach that the examining comprises examining for shadows and/or vessels, wherein the guidance reduces the shadows and/or vessels in the field of view. However, FERRAIOLI teaches avoiding vessels when obtain quality images. “The transducer was positioned in the intercostal space, and measurements were obtained on the best quality image, that is, the one with fewer vessels and the strongest B-mode signal without artifacts.” (p.2248, right column).
XIE is consistent with FERRAIOLI. More specifically, XIE teaches providing guidance to the user to locate the ROIs for measurements. ([0031]). After describing what features are suitable, XIE warns that the ROIs should not otherwise be “located in a region prone to imaging artifacts (e.g., too close or overlapping the boundary between the tissues, near or overlapping vessels or other non-uniform bodily structures).”
It would have been obvious to one having ordinary skill in the art to modify the system to utilize a machined-learned model to examine for shadows and/or vessels and provide guidance that reduces the shadows and/or vessels in the field of view. One having ordinary skill in the art would be motivated to use a machine-learned model to avoid these regions to increase the likelihood that the operator will position the probe accurately and avoid misdiagnoses as taught in XIE. There would have been a reasonable expectation of success because XIE demonstrates that such systems can incorporate AI assistance.
With respect to claim 12 (depending from claim 10), none of LABYED ‘323, FERRAIOLI, KRISHNAN, and AMAR explicitly teach that the examining comprises scoring the field of view for automated placement of the region of interest. NOTE: The recitation “for automated placement of the region of interest” in intended use that does not require that any structure be automatically placed at a region of interest.
However, XIE teaches determining confidence metrics of images in real-time. “During intelligent scan mode, the image data for each acquired frame may be provided to the engine 227 for identification of a suitable view in real-time. As described, the view identification (or view matching) may be performed by a neural network 228, which may include one or any number of stacked, connected or otherwise appropriately arranged networks of artificial neurons. In some examples, the neural network 228 may include a deep convolutional network configured to output, for each input image, a confidence metric (also referred to herein as matching score). The confidence metric (or matching score) may provide an indication of a probability or confidence level that the given image corresponds to the desired or target image view.” ([0050]).
It would have been obvious to one having ordinary skill in the art to modify the system to score the field of view for automated placement of the region of interest. One having ordinary skill in the art would be motivated to use a machine-learned model to help a user or system identify locations that have a high confidence metric so that the probe will be positioned accurately and avoid misdiagnoses as taught in XIE. There would have been a reasonable expectation of success because XIE demonstrates that such systems can incorporate AI assistance.
Claim 13 is rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent Appl. Publ. No. 2018/0289323 A1 to Labyed (hereinafter “LABYED ‘323”) and Ferraioli, Giovanna, et al. “Liver fat quantification with ultrasound: depth dependence of attenuation coefficient.” Journal of Ultrasound in Medicine 42.10 (2023): 2247-2255 (hereinafter “FERRAIOLI”) and/or Krishnan KB, Nagaraj N, Singhal N, Thapar S, Yadav K. A two-parameter model for ultrasonic tissue characterization with harmonic imaging. arXiv preprint arXiv:1712.03495. 2017 Dec 10. (hereinafter “KRISHNAN”) and in further view of U.S. Patent Appl. Publ. No. 2023/0100912 A1 to Amar (hereinafter “AMAR”) as applied to claim 1 above, and further in view of U.S. Patent Appl. Publ. No. 2021/0145409 A1 to Labyed (hereinafter referred to as “LABYED ‘409”).
With respect to claim 13, none of LABYED ‘323, FERRAIOLI, KRISHNAN, and AMAR explicitly teach determining the estimated fat fraction as a field of estimated fat fraction values distributed in a region of interest in the liver, and wherein displaying comprises displaying an ultrasound image with the region of interest coded by the estimated fat fraction values.
In the same field of endeavor, LABYED ‘409 teaches “[f]or parametric ultrasound imaging with an ultrasound scanner, the values for multiple parameters are determined for tissue of a patient using ultrasound.” (Abstract). “FIG. 2 shows an example with shear wave speed map 20B, fat fraction map 20C, and inflammation map 20D shown separately. A B-mode image 20A is also shown in one of the quadrants. The shear wave speed map 20B, fat fraction map 20C, and inflammation map 20D are shown as color overlays in the region of interest 22 where the rest of the spatial representation of tissue is a repeat of the B-mode image 20A.” ([0036]). “The image represents the spatial distribution for each of the multiple parameters. Different values for a
PNG
media_image2.png
381
613
media_image2.png
Greyscale
given parameters may be provided for different locations in the region of interest 22 or across the image.” ([0037]).
Accordingly, LABYED ‘265 teaches a field of estimated fat fraction values (i.e., different values for a given parameter) distributed in a region of interest (for different locations in the region of interest 22), and wherein displaying comprises displaying an ultrasound image with the region of interest coded by the estimated fat fraction values (see Figure 2 in which the region of interest 22 overlays the B-mode image and has different colors based on the values).
It would have been obvious to one having ordinary skill in the art to modify the system to show a field of estimated fat fraction values (differentiated by color) within the ROI and with the ROI overlaying the ultrasound image as taught in LABYED ‘409. One having ordinary skill in the art would be motivated to this image arrangement in order to convey different types of information to the user. There would have been a reasonable expectation of success because LABYED ‘409 demonstrates that such image arrangements can be implemented.
RESPONSE TO APPLICANT’S ARGUMENTS
Applicant’s arguments with respect to the previous Section 103 rejections of claims 1-13 have been 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.
Applicant argues that one having ordinary skill in the art would simply feed “the mathematically compensated BSC or AC…as an input into the machine-learned model….” Examiner disagrees that one would not think to use measurements of the intervening tissue as inputs as LABYED ‘323 teaches training the machine-learning model to learn “what weighted combination (e.g., convolution using learnt kernel) of input values indicates the output.” ([0051]) and KRISHNAN teaches correcting for signal losses based on a thickness of a subcutaneous layer.
Nonetheless, Examiner is relying upon AMAR for teaching that an input to the machine-learned model can be a measurement (e.g., thickness among other things) of the intervening tissue and the output would be based at least in part on the measurement.
Prior Art of Record
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
US-20210196233-A1 describes systems and methods for driving all transducer elements in a transducer array with ultrasound parameters that account for transducer geometric errors and/or beam aberrations resulting from intervening tissues so as to create a high-quality focus at one or more target regions. ([0010]).
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JASON P GROSS whose telephone number is (571)272-1386. The examiner can normally be reached Monday-Friday 9:00-5:00CT.
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, Anne M. Kozak can be reached at (571) 270-5284. 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.
/JASON P GROSS/Examiner, Art Unit 3797
/SERKAN AKAR/Primary Examiner, Art Unit 3797
Prosecution Insights