DETAILED ACTION
Summary
Claims 1-8, and 10-20 are pending in the application. Claims 1-8, and 10-20 are rejected under 35 USC 101. Claims 1-8, and 10-20 are rejected under 35 USC 103.
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 .
Claim Objections
Claim 4 objected to because of the following informalities:
Claim 4 recites “the set of one or more images” in lines 3-4. It should recite “the set of one or more ultrasound image”.
Appropriate correction is required.
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 1-20 rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The claim(s) recite(s) a mental process. This judicial exception is not integrated into a practical application because the additional limitations are generic computer components that are just applying the abstract idea. The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception because performing the abstract idea with generic computer components does not amount to significantly more than the judicial exception.
Claim 1 recites an abstract idea without significantly more. The steps of “a prediction of a presence and/or position of each of a plurality of anatomical structures in the set of one or more ultrasound image of the torso, and a prediction of a presence of the pathological condition and/or a quantitative indication of a severity of the pathological condition; and in response to determining, based on the prediction of the presence and/or position of the plurality of anatomical structures, that the set of one or more ultrasound image is suitable for detecting the pathological condition” are directed to an abstract of a mental process. A user, looking at a set of ultrasound images, and predict the presence of an anatomical structure in the image using only their mind (e.g. by thinking about whether the structure is in the image). A user, looking at the set of ultrasound images, can predict presence of the pathological condition using only their mind (e.g. by thinking about whether the pathology is in the image). A user, thinking about the anatomical features are in the image, can further determine that the image is suitable for detection using only their mind (e.g. by thinking about the image being suitable for detection or not). The step of “generating a ternary decision” is directed to an abstract idea of a mental process. A user, looking at imaging data, can generate a ternary decision using only their mind. The step of training the neural network “using a loss function with a custom regularization term” is directed to a mathematical concept. As detailed in the specification, the loss function is a mathematical algorithm [0130]. The claim recites an abstract idea.
The claim is not integrated into a practical application. The additional limitation of “a memory”, “at least one processor”, and “one or more neural networks” are recited at a high level of generality, and amount to no more than applying the abstract idea using a generic computer system. The step of outputting the prediction is directed to an insignificant post-solution activity of outputting data and does not integrate the abstract idea into a practical application. The claim is directed to an abstract idea. The claim is not significantly more than the judicial exception. As stated above, the additional features are just generic computer components for performing the abstract idea, and an insignificant post-solution activity of displaying data. The claim is not subject matter eligible.
Claim 2 recites an abstract idea without significantly more. The steps of “generate the prediction of the presence and/or position of each of the plurality of anatomical structures in the set of one or more ultrasound image of the torso, and the prediction of the presence of the pathological condition and/or the quantitative indication of the severity of the pathological condition” are directed to abstract idea of a mental process. A user, looking at a set of ultrasound images, and predict the presence of an anatomical structure in the image using only their mind (e.g. by thinking about whether the structure is in the image). A user, looking at the set of ultrasound images, can predict presence of the pathological condition using only their mind (e.g. by thinking about whether the pathology is in the image). The claim recites an abstract idea. The claim is not integrated into a practical application. The additional limitation of “the one or more neural networks comprises a single neural network” is recited at a high level of generality, and amounts to no more than applying the abstract idea using a generic computer system. The claim is not significantly more than the judicial exception. As stated above, the additional features are just generic computer components for performing the abstract idea. The claim is not subject matter eligible.
Claim 3 recites an abstract idea without significantly more. The steps of “generate the prediction of the presence and/or position of each of the plurality of anatomical structures in the set of one or more ultrasound image of the torso”, and “the prediction of the presence of the pathological condition and/or the quantitative indication of the severity of the pathological condition” are directed to abstract idea of a mental process. A user, looking at a set of ultrasound images, can predict the presence of an anatomical structure in the image using only their mind (e.g. by thinking about whether the structure is in the image). A user, looking at the set of ultrasound images, can predict presence of the pathological condition using only their mind (e.g. by thinking about whether the pathology is in the image). The claim recites an abstract idea. The claim is not integrated into a practical application. The additional limitation of “the one or more neural networks comprises a first neural network… and a second, different neural network” is recited at a high level of generality, and amounts to no more than applying the abstract idea using a generic computer system. The claim is not significantly more than the judicial exception. As stated above, the additional features are just generic computer components for performing the abstract idea. The claim is not subject matter eligible.
Claim 4 recites an abstract idea without significantly more. The step of “processing the set of one or more ultrasound image of the torso… only if… it is determined that the set of one or more ultrasound image of the torso is suitable for detecting the pathological condition” is directed to abstract idea of a mental process. A user can, looking at the images, determine the presence of a pathological condition in response to the user deciding the images are suitable using only their mind. The claim recites an abstract idea. The claim is not integrated into a practical application. The additional limitation of “the second, different neural network” is recited at a high level of generality, and amounts to no more than applying the abstract idea using a generic computer system. The claim is not significantly more than the judicial exception. As stated above, the additional features are just generic computer components for performing the abstract idea. The claim is not subject matter eligible.
Claim 5 recites an abstract idea without significantly more. The step of “to process the set of one or more ultrasound image of the torso in conjunction with data generated from the prediction of the presence and/or position of each of the plurality of anatomical structures in the set of one or more ultrasound image of the torso” is directed to abstract idea of a mental process. A user can, looking at the images and thinking about the prediction pertaining to the plurality of structures, determine the presence of a pathological condition in response to the user deciding the images are suitable using only their mind. The claim recites an abstract idea. The claim is not integrated into a practical application. The additional limitation of “the second, different neural network” is recited at a high level of generality, and amounts to no more than applying the abstract idea using a generic computer system. The claim is not significantly more than the judicial exception. As stated above, the additional features are just generic computer components for performing the abstract idea. The claim is not subject matter eligible.
Claim 6 recites an abstract idea without significantly more. The step of “the prediction of the presence of each of the plurality of anatomical structures in the set of one or more ultrasound image of the torso and the prediction of the presence of each of the plurality of anatomical structures comprises a multi-element vector that comprises an element for each of the plurality of anatomical structures that indicates the prediction of the presence of that anatomical structure in the set of one or more ultrasound image of the torso” is directed to an abstract idea of a mental process. A user, looking at the ultrasound can predict the presence of anatomical structures as a multi-element vector using only their mind (e.g. think about a vector which represents the presence of the anatomical features). The claim recites an abstract idea. The claim is not integrated into a practical application. The additional limitation of “a classifier neural network” is recited at a high level of generality, and amounts to no more than applying the abstract idea using a generic computer system. The claim is not significantly more than the judicial exception. As stated above, the additional features are just generic computer components for performing the abstract idea. The claim is not subject matter eligible.
Claim 7 recites an abstract idea without significantly more. The step of “generate the prediction of the position of each of the plurality of anatomical structures in the set of one or more ultrasound image of the torso” is directed to an abstract idea of a mental process. A user, looking at the ultrasound can predict the presence of anatomical structures only their mind. The claim recites an abstract idea. The claim is not integrated into a practical application. The additional limitation of “an object detection neural network” is recited at a high level of generality, and amounts to no more than applying the abstract idea using a generic computer system. The claim is not significantly more than the judicial exception. As stated above, the additional features are just generic computer components for performing the abstract idea. The claim is not subject matter eligible.
Claim 8 recites an abstract idea without significantly more. The step of “generate the prediction of the position of each of the plurality of anatomical structures in the set of one or more ultrasound image of the torso” is directed to an abstract idea of a mental process. A user, looking at the ultrasound can predict the presence of anatomical structures only their mind. The claim recites an abstract idea. The claim is not integrated into a practical application. The additional limitation of “a segmentation neural network” is recited at a high level of generality, and amounts to no more than applying the abstract idea using a generic computer system. The claim is not significantly more than the judicial exception. As stated above, the additional features are just generic computer components for performing the abstract idea. The claim is not subject matter eligible.
Claim 10 recites an abstract idea without significantly more. The steps of processing the plurality of images as a whole and processing each ultrasound image to generate the prediction are directed to an abstract idea of a mental process. A user, looking at the set of images, can predict the presence of an anatomical feature using the images as a whole, or each image, using only their mind. The claim recites an abstract idea. The claim is not integrated into a practical application. The additional limitation of “one or more neural networks” is recited at a high level of generality, and amounts to no more than applying the abstract idea using a generic computer system. The claim is directed to an abstract idea. The claim is not significantly more than the judicial exception. As stated above, the additional features are just generic computer components for performing the abstract idea, and an insignificant post-solution activity of displaying data. The claim is not subject matter eligible.
Claim 11 recites an abstract idea without significantly more. The step of “generate the prediction of the position” is directed to an abstract idea of a mental process. A user, looking at the ultrasound, can predict the presence of anatomical structures only their mind as a single value or a plurality of values. The claim recites an abstract idea. The claim is not integrated into a practical application. The additional limitation of “a classifier neural network” is recited at a high level of generality, and amounts to no more than applying the abstract idea using a generic computer system. The claim is not significantly more than the judicial exception. As stated above, the additional features are just generic computer components for performing the abstract idea. The claim is not subject matter eligible.
Claim 12 recites an abstract idea without significantly more. The step of “generate the quantitative indication of the severity of the pathological condition” is directed to an abstract idea of a mental process. A user, looking at the ultrasound can think of the quantitative indication (e.g. the user can think about which pixels correspond to the pathological condition, and thus generate the segmentation mask). The claim recites an abstract idea. The claim is not integrated into a practical application. The additional limitation of “a segmentation neural network” is recited at a high level of generality, and amounts to no more than applying the abstract idea using a generic computer system. The claim is not significantly more than the judicial exception. As stated above, the additional features are just generic computer components for performing the abstract idea. The claim is not subject matter eligible.
Claim 13 recites an abstract idea without significantly more. The steps of “generate the quantitative indication of the severity of the pathological condition”, “identify bounding boxes of continuous segments in the set of one or more ultrasound image of the torso that correspond to the pathological condition”, and “a model configured to generate a segmentation mask for each identified bounding box that indicates which pixels of that bounding box correspond to the pathological condition” are directed to a mental process. The user can, looking at the ultrasound images, think about a quantitative indication of the severity of the pathological condition, think about bounding boxes on continuous segments of the image, and use a mental model to segment pixels which correspond to the pathological condition. The claim recites an abstract idea. The claim is not integrated into a practical application. The additional limitation of “an object detection neural network” is recited at a high level of generality, and amounts to no more than applying the abstract idea using a generic computer system. The claim is not significantly more than the judicial exception. As stated above, the additional features are just generic computer components for performing the abstract idea. The claim is not subject matter eligible.
Claim 14 recites an abstract idea without significantly more. The steps of “generate the quantitative indication of the severity of the pathological condition”, “generate a segmentation mask”, and “a model configured to determine an area” are directed to a mental process. The user can, looking at the ultrasound images, think about a quantitative indication of the severity of the pathological condition, think the different areas (i.e. segments) in the image, and use a mental model to determine an area of the pathological condition. The claim recites an abstract idea. The claim is not integrated into a practical application. The additional limitation of “a segmentation neural network” is recited at a high level of generality, and amounts to no more than applying the abstract idea using a generic computer system. The claim is not significantly more than the judicial exception. As stated above, the additional features are just generic computer components for performing the abstract idea. The claim is not subject matter eligible.
Claim 15 recites an abstract idea without significantly more. The step of confining the pixels of the segmentation mask to an ultrasound beam is directed to a mental process. A user, looking at the ultrasound image, can disregard (i.e. mask) the pixels which would be outside the beam using on their mind. The claim recites an abstract idea. The claim does not recite any features which either integrate the abstract idea into a practical application or amount to significantly more than the judicial exception. The claim is not subject matter eligible.
Claim 16 recites an abstract idea without significantly more. The steps of a “feature extractor configured to generate the quantitative indication of the severity of the pathological condition, wherein the quantitative indication of the severity of the pathological condition comprises a value that represents a predicted severity of the pathological condition” is a limitation that can reasonably be performed in the human mind. A user, looking at an ultrasound image, can extract (i.e. think about) features and determine the severity using only their mind. The claim recites an abstract idea. The claim is not integrated into a practical application. The additional limitation of “the one or more neural networks” is recited at a high level of generality, and amounts to no more than applying the abstract idea using a generic computer system. The claim is not significantly more than the judicial exception. As stated above, the additional features are just generic computer components for performing the abstract idea. The claim is not subject matter eligible.
Claim 17 recites an abstract idea without significantly more. The steps of processing the plurality of images as a whole and processing each ultrasound image to generate the prediction are directed to an abstract idea of a mental process. A user, looking at the set of images, can predict the presence of an anatomical feature using the images as a whole, or each image individually, using only their mind. The claim recites an abstract idea. The claim is not integrated into a practical application. The additional limitation of “one or more neural networks” is recited at a high level of generality, and amounts to no more than applying the abstract idea using a generic computer system. The claim is directed to an abstract idea. The claim is not significantly more than the judicial exception. As stated above, the additional features are just generic computer components for performing the abstract idea, and an insignificant post-solution activity of displaying data. The claim is not subject matter eligible.
Claim 18 is further narrowing the abstract idea of predicting a pathological condition and is also directed to an abstract idea of a mental process. A user, looking at the ultrasound images, can predict the pathological condition of free fluid in the tissue. The claim recites an abstract idea. The claim contains no limitations that either integrate the abstract idea into a practical application or amount to significantly more than the judicial exception. The claim is not subject matter eligible.
Claim 19 recites an abstract idea without significantly more. The step “the quantitative indication of the severity of the pathological condition is an estimate of a quantity of the free fluid in the torso” is to an abstract idea of a mental process. A user, looking at the ultrasound images, can estimate a quantity of the free fluid in the torso. The claim recites an abstract idea. The claim contains no limitations that either integrate the abstract idea into a practical application or amount to significantly more than the judicial exception. The claim is not subject matter eligible.
Claim 20 recites an abstract idea without significantly more. The steps of “a prediction of a presence and/or position of each of a plurality of anatomical structures in the set of one or more ultrasound image of the torso, and a prediction of a presence of the pathological condition and/or a quantitative indication of a severity of the pathological condition; and in response to determining, based on the prediction of the presence and/or position of the plurality of anatomical structures, that the set of one or more ultrasound image is suitable for detecting the pathological condition” are directed to an abstract of a mental process. A user, looking at a set of ultrasound images, and predict the presence of an anatomical structure in the image using only their mind (e.g. by thinking about whether the structure is in the image). A user, looking at the set of ultrasound images, can predict presence of the pathological condition using only their mind (e.g. by thinking about whether the pathology is in the image). A user, thinking about the anatomical features are in the image, can further determine that the image is suitable for detection using only their mind (e.g. by thinking about the image being suitable for detection or not). The steps of “determining… whether the set is suitable”, “generating a ternary decision” are directed to an abstract idea of a mental process. A user, looking at imaging data, can generate a binary/ternary decision using on their mind. The step of training the neural network “using a loss function with a custom regularization term” is directed to a mathematical concept. As detailed in the specification, the loss function is a mathematical algorithm [0130]. The claim recites an abstract idea.
The claim is not integrated into a practical application. The additional limitation of “one or more processors”, and “one or more neural networks” are recited at a high level of generality, and amount to no more than applying the abstract idea using a generic computer system. The step of outputting the prediction is directed to an insignificant post-solution activity of outputting data and does not integrate the abstract idea into a practical application. The claim is directed to an abstract idea. The claim is not significantly more than the judicial exception. As stated above, the additional features are just generic computer components for performing the abstract idea, and an insignificant post-solution activity of displaying data. The claim is not subject matter eligible.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1-2, 6-8, 10-11, 17-18, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Callcut et al. (U.S PGPub 2022/0293243 A1) in view of Fleurentin et al. (WO 2024/056896 A1), Xie et al. (U.S PGPub 2021/0177373 A1), Lilaonitkul et al. (U.S PGPub 2024/0185428 A1), and Rothberg et al. (U.S PGPub 2017/0360412 A1).
Regarding Claim 1, Callcut teaches a system to detect a pathological condition based on a set of one or more ultrasound image of a torso (Abstract), the system comprising:
a memory storing instructions [0048]; and
at least one processor coupled to the memory [0048], the at least one processor configured to execute the instructions [0047]-[0048] to perform a method comprising:
processing the set of one or more ultrasound image of the torso using one or more neural networks [0032]+[0067] to generate:
a prediction of a presence and/or position of each of a plurality of anatomical structures in the set of one or more ultrasound image of the torso [0062], and
a prediction of a presence of the pathological condition and/or a quantitative indication of a severity of the pathological condition [0062] (the presence of free fluid indicates a pathological condition [0007]; and
outputting the prediction of the presence of the pathological condition and/or the quantitative indication of the severity of the pathological condition [0062]+[0074].
Callcut fails to explicitly teach in response to determining, based on the prediction of the presence and/or position of the plurality of anatomical structures, that the set of one or more ultrasound image is suitable for detecting the pathological condition. Callcut fails to explicitly teach wherein generating, based on the prediction of the presence and/or position of the plurality of anatomical structures, whether the set of one or more ultrasound image of the torso is suitable for detecting the pathological condition comprises: generating a ternary decision as to whether the set of one or more ultrasound image of the torso is suitable for detecting the pathological condition, the ternary decision indicating whether the set of one or more ultrasound image of the torso is (i) suitable for detecting the pathological condition, (ii) not suitable for detecting the pathological condition, or (iii) comprises an anatomically impossible combination of the plurality of anatomical structures.
Fleurentin teaches a system for determining whether an ultrasonic image is a target image (Abstract). This system determines whether an image is of the target anatomical structure (i.e. is a suitable image) before further processing the images (Pg. 2, lines 14-27). The classes are based on the presence/position of anatomical structures (Pg. 14, lines 8-22).
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the system of Callcut to determine that the ultrasound image is suitable, as taught by Fleurentin, because this allows the system to only notify the user about the diagnosis when the diagnosis is actually relevant, as recognized by Fleurentin (Pg. 2, lines 19-27), thereby simplifying the interpretation of the ultrasound images and allowing for more accurate interpretation of the images.
The combination fails to explicitly teach generating a decision as to whether the set of one or more ultrasound image of the torso is suitable for detecting the pathological condition.
Xie teaches a system for automatically identifying features in ultrasound images (Abstract). This system generates a decision as to whether the set of one or more images is suitable for detecting the pathological condition [0051]-[0052] (the “match” categories are suitable for detecting the pathological condition).
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the combined system to generate a binary decision, as taught by Xie, because this allows for the proper view of the anatomical structures to be used for diagnosing the patient, thereby simplifying the procedure and improving the accuracy of the diagnostic procedure, as recognized by Xie [0002]-[0003].
The combination fails to explicitly teach that the decision is a ternary decision, which further includes the determination that the image iii) comprises an anatomically impossible combination of the plurality of anatomical structures.
Lilaonitkul teaches a system for automatically determining structures in medical image data (Abstract). This system contains a sub-network, which determines whether the anatomical structures are in an anatomically impossible configuration [0056].
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the combined system to make the decision a ternary decision and further determine whether the image is anatomically impossible, as taught by Lilaonitkul, because this improves image segmentation/analysis, and makes the image analysis more robust even when there is image deformation, as recognized by Lilaonitkul [0056].
The combination fails to explicitly teach wherein the one or more neural networks is trained to generate the prediction of the presence and/or position of each of the plurality of anatomical structures on ultrasound images of a particular ultrasound mode with annotations for the presence of each of the plurality of anatomical structures using a loss function with a custom regularization term integrated therein that constrains a prediction of the presence and/or position of each of the plurality of anatomical structures generated during training to plausible combinations of the plurality of anatomical structures.
Rothberg teaches a system for analyzing ultrasound images (Abstract). This system wherein the one or more neural networks is trained [0006] to generate the prediction of the presence and/or position of each of the plurality of anatomical structures on ultrasound images of a particular ultrasound mode with annotations for the presence of each of the plurality of anatomical structures [0006]+[0283]-[0284] (The landmarks are indicative of the presence of anatomical structures) using a loss function with a custom regularization term integrated therein [0294] (defined loss is a loss function) that constrains a prediction of the presence and/or position of each of the plurality of anatomical structures generated during training to plausible combinations of the plurality of anatomical structures [0285]+[0298] (“legal shapes” are plausible combinations of structures).
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the combined system to train the neural network, as taught by Rothberg, because this allows the ultrasound images to be accurately classified [0261], thereby allowing the ultrasound image procedure to be performed more easily, and effectively, by technicians with less training, as recognized by Rothberg [0003]+[0005]-[0006]
Regarding Claim 2, the combination of references teaches the invention substantially as claimed. Callcut further teaches wherein the one or more neural networks comprises a single neural network configured to generate the prediction of the presence and/or position of each of the plurality of anatomical structures in the set of one or more ultrasound image of the torso, and the prediction of the presence of the pathological condition and/or the quantitative indication of the severity of the pathological condition [0062]+[0067] (the same CNN predicts the presence of the structure (e.g. by segment and predicts the presence of the pathological condition (e.g. by segmenting the free fluid).
Regarding Claim 6, the combination of references teaches the invention substantially as claimed. Callcut further teaches wherein the one or more neural networks comprises a classifier neural network configured to generate the prediction of the presence of each of the plurality of anatomical structures in the set of one or more ultrasound image of the torso [0031]+[0168]-[0169] (as the neural network outputs a classification, it is a classifier neural network).
Callcut fails to explicitly teach the prediction of the presence of each of the plurality of anatomical structures comprises a multi-element vector that comprises an element for each of the plurality of anatomical structures that indicates the prediction of the presence of that anatomical structure in the set of one or more ultrasound image of the torso.
Xie teaches a system for automatically identifying features in an ultrasound images (Abstract). This system outputs a prediction of the presence of each of the plurality of anatomical structures comprises a multi-element vector that comprises an element for each of the plurality of anatomical structures that indicates the prediction of the presence of that anatomical structure in the set of one or more ultrasound image of the torso [0051]+[0053]+[0055] (output vector).
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the system of Callcut to output the prediction of the structures as a multi-element vector, as taught by Xie, as the substitution for one known method for outputting the results of a classifier neural network with another yields predictable results to one of ordinary skill in the art. One of ordinary skill would have been able to carry out such a substitution, and the results of outputting the predictions of the anatomical structures of Callcut in multi-element vector of Xie are reasonably predictable.
Regarding Claim 7, the combination of references teaches the invention substantially as claimed. Callcut further teaches wherein the one or more neural networks comprises an object detection neural network configured to generate the prediction of the position of each of the plurality of anatomical structures in the set of one or more ultrasound image of the torso [0062]+[0067] (as the neural network is detecting objects, it is considered an object detection neural network).
Regarding Claim 8, the combination of references teaches the invention substantially as claimed. Callcut further teach wherein the one or more neural networks comprises a segmentation neural network configured to generate the prediction of the position of each of the plurality of anatomical structures in the set of one or more ultrasound image of the torso [0062]+[0067] (as the software performs segmentation, and the software is a neural network, it is therefore a segmentation neural network).
Regarding Claim 10, the combination of references teaches the invention substantially as claimed. Callcut further teaches wherein the set of one or more ultrasound image of the torso comprises a plurality of ultrasound images of the torso [0062]+[0169] and processing the set of one or more ultrasound image of the torso using the one or more neural networks to generate the prediction of the presence and/or position of each of the plurality of anatomical structures in the set of one or more ultrasound image [0062] comprises:
processing the plurality of ultrasound images of the torso as a whole using a neural network of the one or more neural networks to generate the prediction of the presence and/or position of each of the plurality of anatomical structures [0062]-[0063]+[0169] (the system analyzes all of the images, and as such it can be considered that the images are analyzed “as a whole”); or
processing each ultrasound image of the plurality of ultrasound images of the torso using a neural network of the one or more neural networks to generate a prediction of the presence and/or position of each of the plurality of anatomical structures in that ultrasound image [0044]+[0062], and generating the prediction of the presence and/or position of each of the plurality of anatomical structures in the set of one or more ultrasound image of the torso based on the prediction of the presence and/or position of that anatomical structure for each ultrasound image of the plurality of ultrasound images of the torso [0044]+[0062]-[0063]+[0174].
Regarding Claim 11, the combination of references teaches the invention substantially as claimed. Callcut further teaches wherein the one or more neural networks comprises a classifier neural network configured to generate the prediction of the presence of the pathological condition [0158] (as the neural network outputs a classification, it is a classifier neural network), and the prediction of the presence of the pathological condition comprises (i) a single value that indicates whether the pathological condition is predicted to be present; or (ii) a plurality of values that indicate varying degrees of detection of the pathological condition [0158] (a binary output is a single value that indicates the pathological condition is present).
Regarding Claim 17, the combination of references teaches the invention substantially as claimed. Callcut further teaches wherein the set of one or more ultrasound image of the torso comprises a plurality of ultrasound images of the torso [0062]+[0159] and a neural network of the one or more neural networks [0062] is configured to:
process the plurality of ultrasound images of the torso as a whole to generate the prediction of the presence of the pathological condition and/or the quantitative indication of the severity of the pathological condition [0062]-[0063]+[0159]+0169] (the system analyzes all of the images, and as such it can be considered that the images are analyzed “as a whole”); or
process each ultrasound image of the plurality of ultrasound images of the torso separately to generate a prediction of the presence of the pathological condition and/or the quantitative indication of the severity of the pathological condition for that ultrasound image [0044]+[0062], and generate the prediction of the presence of the pathological condition and/or the quantitative indication of the severity of the pathological condition based on the prediction of the presence of the pathological condition and/or the quantitative indication of the severity of the pathological condition for each ultrasound image of the plurality of ultrasound images of the torso [0044]+[0062]-[0063]+[0174].
Regarding Claim 18, the combination of references teaches the invention substantially as claimed. Callcut further teaches wherein the pathological condition is free fluid in the torso [0007].
Regarding Claim 20, Callcut teaches a method for detecting a pathological condition based on a set of one or more ultrasound image of a torso (Abstract), the method comprising, at one or more processors [0048]:
processing the set of one or more ultrasound image of the torso using one or more neural networks [0032]+[0067] to generate:
a prediction of a presence and/or position of each of a plurality of anatomical structures in the set of one or more ultrasound image of the torso [0062], and
a prediction of a presence of the pathological condition and/or a quantitative indication of a severity of the pathological condition [0062] (the presence of free fluid indicates a pathological condition [0007]; and
outputting the prediction of the presence of the pathological condition and/or the quantitative indication of the severity of the pathological condition [0062]+[0074], otherwise not outputting the prediction of the presence of the pathological condition and/or the quantitative indication of the severity of the pathological condition (this is contingent on the images being found not suitable. As the images were found suitable, the contingency was not met and this limitation is not required by the claims (MPEP 2111.04).
Callcut fails to explicitly teach in response to determining, based on the prediction of the presence and/or position of the plurality of anatomical structures, that the set of one or more ultrasound image is suitable for detecting the pathological condition. Callcut fails to explicitly teach wherein generating, based on the prediction of the presence and/or position of the plurality of anatomical structures, a ternary decision indicating whether the set of one or more ultrasound image of the torso is (i) suitable for detecting the pathological condition, (ii) not suitable for detecting the pathological condition, or (iii) comprises an anatomically impossible combination of the plurality of anatomical structures.
Fleurentin teaches a system for determining whether an ultrasonic image is a target image (Abstract). This system determines whether an image is of the target anatomical structure (i.e. is a suitable image) before further processing the images (Pg. 2, lines 14-27). The classes are based on the presence/position of anatomical structures (Pg. 14, lines 8-22).
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the system of Callcut to determine that the ultrasound image is suitable, as taught by Fleurentin, because this allows the system to only notify the user about the diagnosis when the diagnosis is actually relevant, as recognized by Fleurentin (Pg. 2, lines 19-27), thereby simplifying the interpretation of the ultrasound images and allowing for more accurate interpretation of the images.
The combination fails to explicitly teach generating a decision as to whether the set of one or more ultrasound image of the torso is suitable for detecting the pathological condition.
Xie teaches a system for automatically identifying features in ultrasound images (Abstract). This system generates a decision as to whether the set of one or more images is suitable for detecting the pathological condition [0051]-[0052] (the “match” categories are suitable for detecting the pathological condition).
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the combined system to generate a decision, as taught by Xie, because this allows for the proper view of the anatomical structures to be used for diagnosing the patient, thereby simplifying the procedure and improving the accuracy of the diagnostic procedure, as recognized by Xie [0002]-[0003].
The combination fails to explicitly teach that the decision is a ternary decision, which further includes the determination that the image iii) comprises an anatomically impossible combination of the plurality of anatomical structures.
Lilaonitkul teaches a system for automatically determining structures in medical image data (Abstract). This system contains a sub-network, which determines whether the anatomical structures are in an anatomically impossible configuration [0056].
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the combined system to make the decision a ternary decision and further determine whether the image is anatomically impossible, as taught by Lilaonitkul, because this improves image segmentation/analysis, and makes the image analysis more robust even when there is image deformation, as recognized by Lilaonitkul [0056].
The combination fails to explicitly teach wherein the one or more neural networks is trained to generate the prediction of the presence and/or position of each of the plurality of anatomical structures on ultrasound images of a particular ultrasound mode with annotations for the presence of each of the plurality of anatomical structures using a loss function with a custom regularization term integrated therein that constrains a prediction of the presence and/or position of each of the plurality of anatomical structures generated during training to plausible combinations of the plurality of anatomical structures.
Rothberg teaches a system for analyzing ultrasound images (Abstract). This system wherein the one or more neural networks is trained [0006] to generate the prediction of the presence and/or position of each of the plurality of anatomical structures on ultrasound images of a particular ultrasound mode with annotations for the presence of each of the plurality of anatomical structures [0006]+[0283]-[0284] (The landmarks are indicative of the presence of anatomical structures) using a loss function with a custom regularization term integrated therein [0294] (defined loss is a loss function) that constrains a prediction of the presence and/or position of each of the plurality of anatomical structures generated during training to plausible combinations of the plurality of anatomical structures [0285]+[0298] (“legal shapes” are plausible combinations of structures).
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the combined system to train the neural network, as taught by Rothberg, because this allows the ultrasound images to be accurately classified [0261], thereby allowing the ultrasound image procedure to be performed more easily, and effectively, by technicians with less training, as recognized by Rothberg [0003]+[0005]-[0006]
Claims 3-5 are rejected under 35 U.S.C. 103 as being unpatentable over Callcut in view of Fleurentin, Xie, Lilaonitkul, and Rothberg as applied to claim 1 above, and further in view of Xue et al. (CN116934683A).
Regarding Claim 3, the combination of references teaches the invention substantially as claimed. Callcut fails to explicitly teach wherein the one or more neural networks comprises a first neural network configured to generate the prediction of the presence and/or position of each of the plurality of anatomical structures in the set of one or more ultrasound image of the torso and a second, different neural network configured to generate the prediction of the presence of the pathological condition and/or the quantitative indication of the severity of the pathological condition.
Xue teaches an artificial intelligence system for ultrasound images [0001]. This system comprises a first neural network configured to generate the prediction of the presence and/or position of each of the plurality of anatomical structures in the set of one or more ultrasound image of the torso [0022]+[0033]+[0037]. The system contains a second, different neural network configured to generate the prediction of the presence of the pathological condition and/or the quantitative indication of the severity of the pathological condition [0023]+[0038]-0039].
It would have been obvious to one of ordinary skill in the art to use two neural networks to respectively generate a presence/position of anatomical structures and presence/severity of the pathological condition, as taught by Xue, because this improves the accuracy of trauma classification, as recognized by Xue [0043].
Regarding Claim 4, the combination of references teaches the invention substantially as claimed. Callcut fails to explicitly teach wherein the method further comprises processing the set of one or more ultrasound image of the torso using the second, different neural network only if, after processing the set of one or more images using the first neural network, it is determining that the set of one or more ultrasound image of the torso is suitable for detecting the pathological condition.
Fleurentin further teaches only processing the set of ultrasound images in response to determined that the set one or more images is suitable (pg. 2, lines 19-27).
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the system of Callcut to determine that the ultrasound image is suitable, as taught by Fleurentin, because this allows the system to only notify the user about the diagnosis when the diagnosis is actually relevant, as recognized by Fleurentin (Pg. 2, lines 19-27), thereby simplifying the interpretation of the ultrasound images and allowing for more accurate interpretation of the images.
The combination is silent regarding the second, different neural network.
Xue teaches a second, different neural network configured to generate the prediction of the presence of the pathological condition and/or the quantitative indication of the severity of the pathological condition [0023]+[0038]-[0039].,
It would have been obvious to one of ordinary skill in the art to use two neural networks to respectively generate a presence/position of anatomical structures and presence/severity of the pathological condition, as taught by Xue, because this improves the accuracy of trauma classification, as recognized by Xue [0043].
Regarding Claim 5, the combination of references teaches the invention substantially as claimed. Callcut fails to explicitly teach wherein the second, different neural network is configured to process the set of one or more ultrasound image of the torso in conjunction with data generated from the prediction of the presence and/or position of each of the plurality of anatomical structures in the set of one or more ultrasound image of the torso.
Xue further teaches wherein the second, different neural network is configured to process the set of one or more ultrasound image of the torso in conjunction with data generated from the prediction of the presence and/or position of each of the plurality of anatomical structures in the set of one or more ultrasound image of the torso [0022]-[0023]+[0038]-[0039] (the segmentation prediction result is the one or more of the images in conjunction with the generated data, as the generated data is necessarily obtained by processing the image).
It would have been obvious to one of ordinary skill in the art to use two neural networks to respectively generate a presence/position of anatomical structures and presence/severity of the pathological condition, as taught by Xue, because this improves the accuracy of trauma classification, as recognized by Xue [0043].
Claims 12-14, 16, and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Callcut in view of Fleurentin, Xie, Lilaonitkul, and Rothberg as applied to claim 1 and 18, respectively, above, and further in view of Ghoshal et al. (U.S PGPub 2023/0316523 A1).
Regarding Claim 12, the combination of references teaches the invention substantially as claimed. Callcut further teaches wherein the one or more neural networks comprises a segmentation neural network [0062]+[0172] for one or more ultrasound image in the set of one or more ultrasound image of the torso [0062], a segmentation mask that indicates which pixels of the ultrasound image correspond to the pathological condition [0062]-[0063]+[0172] (a mask which reflects the segmentation indicates the pixels corresponding to the pathological condition).
Callcut is silent regarding generating a quantitative indication of the severity of the pathological condition.
Ghoshal teaches a system for analyzing ultrasound images (Abstract). This system outputs a quantitative indication of a severity of the amount of free fluid located in the torso [0053]. This can be done by using a neural network to segment the different portions of the anatomy [0032].
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the combined system to output a quantitative indication of the severity of the pathological condition, as taught by Ghoshal, because this assists in image interpretation, thereby improving the quality of care, as recognized by Ghoshal [0003]-[0004].
Regarding Claim 13, the combination of references teaches the invention substantially as claimed. Callcut further teaches that neural network comprises: an object detection neural network [0062]+[0067] (as the neural network is detecting objects, it is considered an object detection neural network) configured to identify bounding boxes of continuous segments in the set of one or more ultrasound image of the torso that correspond to the pathological condition [0062] (the bounding boxes are placed around the free fluid); and a model configured to generate a segmentation mask for each identified bounding box that indicates which pixels of that bounding box correspond to the pathological condition [0062]-[0063] (the system applies the bounding boxes to the free fluid, and also applies a mask to show the free fluid. One of ordinary skill would recognize that the system would contain a model that generates the mask for each of the bounding boxes).
Callcut is silent regarding wherein the one or more neural networks comprises a neural network that is configured to generate the quantitative indication of the severity of the pathological condition.
Ghoshal teaches a system for analyzing ultrasound images (Abstract). This system outputs a quantitative indication of a severity of the amount of free fluid located in the torso [0053]. This can be done by using a neural network to segment the different portions of the anatomy [0032].
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the combined system to output a quantitative indication of the severity of the pathological condition, as taught by Ghoshal, because this assists in image interpretation, thereby improving the quality of care, as recognized by Ghoshal [0003]-[0004].
Regarding Claim 14, the combination of references teaches the invention substantially as claimed. Callcut further teaches a segmentation neural network configured to generate one or more segmentation mask from the one or more ultrasound image of the torso [0062]+[0067] (as the software performs segmentation, and the software is a neural network, it is therefore a segmentation neural network).
Callcut fails to explicitly teach wherein the one or more neural networks comprise a neural network that is configured to generate the quantitative indication of the severity of the pathological condition and model configured to determine an area of the set of one or more ultrasound image of the torso corresponding to the pathological condition based on the one or more segmentation mask.
Ghoshal teaches a system for analyzing ultrasound images (Abstract). This system outputs a quantitative indication of a severity of the amount of free fluid located in the torso [0053]. This can be done by using a neural network to segment the different portions of the anatomy [0032]. This system calculates an area of the free fluid on the ultrasound image [0015]+0050]
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the combined system to output a quantitative indication of the severity of the pathological condition, as taught by Ghoshal, because this assists in image interpretation, thereby improving the quality of care, as recognized by Ghoshal [0003]-[0004]. One of ordinary skill in the art would recognize that, in the combination, the system of Callcut would perform the segmentation mask on the free fluid on the images, and the area of the free fluid would be calculated from that, as taught by Ghoshal.
Regarding Claim 16, the combination of references teaches the invention substantially as claimed. Callcut fails to explicitly teach wherein the one or more neural networks comprises a feature extractor configured to generate the quantitative indication of the severity of the pathological condition, wherein the quantitative indication of the severity of the pathological condition comprises a value that represents a predicted severity of the pathological condition.
Ghoshal teaches a system for analyzing ultrasound (Abstract). This system contains a features extractor in the neural network [0051]. The extracted features are used for generating the quantitative indication of the severity of the pathological condition [0051]. The quantitative indication of the severity of the pathological condition comprises a value that represents a predicted severity of the pathological condition [0053].
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the combined system to output a value of the severity of the pathological condition, as taught by Ghoshal, because this assists in image interpretation, thereby improving the quality of care, as recognized by Ghoshal [0003]-[0004].
Regarding Claim 19, the combination of references teaches the invention substantially as claimed. Callcut fails to explicitly teach wherein the quantitative indication of the severity of the pathological condition is an estimate of a quantity of the free fluid in the torso.
Ghoshal teaches a system for analyzing an ultrasound image (Abstract). This system outputs an indication of the severity of the pathological condition [0051] which is an estimate of the quantity of the free fluid in the torso [0053].
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the combined system to output a value of the severity of the pathological condition, as taught by Ghoshal, because this assists in image interpretation, thereby improving the quality of care, as recognized by Ghoshal [0003]-[0004].
Claim 15 is rejected under 35 U.S.C. 103 as being unpatentable over Callcut in view of Fleurentin, Xie, Lilaonitkul, Rothberg, and Ghoshal as applied to claim 14 above, and further in view of Vanberlo et al. (U.S Patent 12,373,926 B1).
Regarding Claim 15, the combination of references teaches the invention substantially as claimed. The combination fails to explicitly teach wherein each segmentation mask is confined to pixels of a corresponding ultrasound image corresponding to an ultrasound beam.
Vanberlo teaches a system for masking pixels in an ultrasound image (Abstract). This system masks pixels around the ultrasound beam (Fig. 7, 708) (Col 14, lines 15-19).
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the combined system to confine the segmentation mask to an ultrasound image corresponding to the ultrasound beam, as taught by Vanberlo, because this provides a better method for removing artifacts in the image, thereby increasing the quality of the image, as recognized by Vanberlo (Col 2-3, lines 54-3).
Response to Arguments
Applicant's arguments filed 8/13/2026 have been fully considered but they are not persuasive.
Applicant argues that the claims are subject matter eligible. The Applicant states that the 101 rejection is contrary to the policy set by the USPTO director. The Examiner disagrees. The rejections are consistent with Office policy as set forth in the MPEP, which reflects the policy set forth by the director.
The Applicant argues that the claim is not directed to an abstract idea. The Applicant argues that, as the neural network is performed the predictions, the claim cannot be practically performed in the human mind. The Examiner disagrees. The predictions and the generation of the ternary decision are steps which can practically be performed in the human mind, as a doctor, looking at ultrasound images, can identify structures and diagnose pathologic conditions. The neural network is trained using a loss function and custom regularization term. By the broadest reasonably interpretation, these are mathematical algorithms (i.e. a function is a mathematical concept, the regularization term is a part of the algorithm). Therefore, the claimed neural network is directed to a mathematical concept, which is an abstract idea.
The Applicant argues that the claim contains an improvement and is integrated into a practical application. The Applicant argues that the use of neural network using a loss function is and improvements to computer component or system performance based upon adjustments to parameters of a machine learning model associated with tasks or workstreams; Ex parte Desjardins, Appeal No. 2024-000567 (PTAB September 26, 2025, Appeals Review Panel Decision) (precedential)”. The Examiner disagrees. As detailed above, the neural network recites mathematical concepts. “It is important to note, the judicial exception alone cannot provide the improvement. The improvement can be provided by one or more additional elements.” (MPEP 2106.05(a)). In this case, the abstract ideas alone provide the improvement; the improved performance of the neural network is reflected by the mathematical concept, the use of a quality gate is a mental process, and the predictions are also mental processes. The Applicant argues that claims are analogous to Ex parte Desjardins and reflect an improvement in how the machine learning algorithm itself operates. The Examiner disagrees, as the specification does not indicate that the performed steps are for improving the functioning of the machine learning algorithm itself, but rather appear to be necessary steps for training the algorithm to identify the pathological features. Therefore, the claims are not integrated into a practical application.
The Applicant argues that the claims are significantly more than the judicial exception. The Examiner disagrees. As detailed above, the claims recite a judicial exception, and the claimed improvement is reflected solely by the judicial exception, which is not enough to amount to significantly more than the judicial exception. Therefore, the claim is not subject matter eligible and the rejection under 35 USC 101 is maintained.
Applicant’s arguments with respect to claim 1 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. Claim 1 is now rejected under a combination of Callcut, Fleurentin, Xie, Lilaonitkul, and Rothberg. The other claims remain rejected under 35 USC 103 for similar reasons.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SEAN D MATTSON whose telephone number is (408)918-7613. The examiner can normally be reached Monday - Friday 9 AM - 5 PM PST.
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, Pascal Bui-Pho can be reached at (571) 272-2714. 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.
/SEAN D MATTSON/Primary Examiner, Art Unit 3798
Prosecution Insights