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 .
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.
Claim(s) 22 is/are rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter. The claim(s) does/do not fall within at least one of the four categories of patent eligible subject matter.
Claim 22 recites a machine-readable medium. When read in light of the Specification as originally filed, the broadest reasonable interpretation of this limitation would include a propagated signal and/or carrier wave.
Accordingly, the claim is not directed to at least one statutory category of invention.
Claim(s) 1,3-5,8-12,14-19,22-25,27 and 30-32 is/are rejected under 35 U.S.C. 101 because the claimed invention is directed to abstract idea without significantly more.
Claim 1 recites:
A machine learning model implemented on at least one processor and pretrained using a loss function with at least one negative predictive value (NPV) term to make one or more oncological medical predictions using one or both of (a) medical images of a patient or (b)radiomic features, pathomic features, or a combination of the radiomic features and the pathomic features extracted from the medical images of the patient, at least some of the medical images of the patient including at least one lesion, and at least some of the radiomic features or the pathomic features extracted from regions of the medical images of the patient including or around the at least one lesion.
Step 1:
The claim as a whole falls within at least one statutory category, i.e. a process, machine, manufacture, or composition of matter.
Step 2A Prong One:
The highlighted portion, as drafted, is a process that, under its broadest reasonable interpretation, falls under “Mathematical concepts” because the step of pretraining a machine learning model using a loss function is a form of math, i.e. mathematical relationships, mathematical formulas or equations, mathematical calculations. MPEP § 2106.04(a)(2)(I)
The highlighted portion, as drafted, is a process that, under its broadest reasonable interpretation, falls under “Certain methods of organizing human activity” because the step of deriving data from a matrix is traditionally performed by a physician when diagnosing and treating a patient, i.e. managing personal behavior or relationships or interactions between people (including social activities, teaching, and following rules or instructions). MPEP 2106.04(a)(2)(II)
The highlighted portion, as drafted, is a process that, under its broadest reasonable interpretation, falls under “Mental processes”.
But for a generic computer recited with a high level of generality in a post hoc manner to implement the abstract idea, the deriving step may be performed in the human mind either mentally or with pen and paper.
Accordingly, these limitations have been found to be directed towards concepts performed in the human mind (including an observation, evaluation, judgment, opinion). MPEP 2106.04(a)(2)(III)
The different categories of abstract ideas are being considered together as one single abstract idea. MPEP 2106.04(II)(B)
Dependent claim(s) recite(s) additional subject matter which further narrows or defines the abstract idea embodied in the claims (such as claim(s) 3-4, 30-31 reciting limitations further defining the abstract idea, which may be performed in the mind but for recitation of generic computer components, and/or may be a method of managing relationship or interactions between people).
Step 2A Prong Two:
This judicial exception is not integrated into a practical application. In particular, the claim recites the following additional element(s), if any:
A machine learning model implemented on at least one processor.
The additional element(s) do(es) not integrate the abstract idea into a practical application, other than the abstract idea per se.
Regarding the processor, the Specification as originally filed on 15 November 2024 discloses a generic computer (page 17 paragraph 0062). Accordingly, this limitation amount(s) to mere instructions to apply an exception (invoking computers as a tool to perform the abstract idea). MPEP 2106.05(f))
Similarly, the machine learning model has been recited with a high level of generality, and merely amounts to mere instructions to apply an exception (invoking computers as a tool to perform the abstract idea). MPEP 2106.05(f))
Dependent claim(s) recite(s) additional subject matter which amount to limitation(s) consistent with the additional element(s) in the independent claims (such as claim(s) 3-4 reciting using a machine learning model deployed on a generic processor, additional limitation(s) which amount(s) to invoking computers as a tool to perform the abstract idea).
Looking at the limitations as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology. Their collective functions merely provide conventional computer implementation and do not impose a meaningful limit to integrate the abstract idea into a practical application.
Accordingly, the additional elements do not integrate the judicial exception into a practical application because it does not impose any meaningful limits on practicing the abstract idea.
Accordingly, the claim recites an abstract idea.
Step 2B:
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements amount to no more than mere instructions to apply an exception, add insignificant extra-solution activity to the abstract idea, and/or generally link the abstract idea to a particular technological environment or field of use.
The additional elements, as discussed above and incorporated herein, amount to no more than mere instructions to apply an exception, add insignificant extra-solution activity to the abstract idea, and/or generally link the abstract idea to a particular technological environment or field of use, as discussed above and incorporated herein.
Mere instructions to apply an exception, insignificant extra-solution activity, and linking to a particular technological environment using a generic computer component cannot provide an inventive concept.
Dependent claims recite additional subject matter which amount to limitations consistent with the additional elements in the independent claims. MPEP 2106.05(d)(II)(ii))
Looking at the limitations as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology. Their collective functions merely provide conventional computer implementation.
The claim is not patent eligible.
Claim(s) 5,8-12,14-19,22-25,27 and 32 recite(s) substantially similar limitations as those of claim(s) 1, 3-4, 30-31 above, and are therefore rejected for substantially similar rationale as applied above, and incorporated herein.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claim(s) 1,3-5,8-12,14-19,22-25,27 and 30 is/are rejected under 35 U.S.C. 102(a)(1) and (a)(2) as being anticipated by Du (WO2022178329, cited by Applicant on 05 February 2025).
Claim 1: Du discloses:
A machine learning model implemented on at least one processor and pretrained using a loss function (Fig. 3B, para [075] "The tissue-type CNN classifier was trained on the training set separately from the rest of the framework on the tissue type classification task. The tissuetype CNN was trained on a per-slice basis by optimizing a class- weighted categorical cross-entropy loss function with an adaptive stochastic gradient descent-based optimization algorithm, Adam, using a batch size of 512 samples for 500 epochs.") with at least one negative predictive value (NPV) term (Figs. 8-14; para [0149] "Overall accuracy was computed across examples from all classes. Precision is defined as the number of true positives over the number of true and false positives. Recall is defined as the number of true positives over true positives and false negatives.") to make one or more oncological medical predictions using one or both of (a) medical images of a patient (Fig. 2 para [053] "The present disclosure provides a deep learning (DL) framework to classify medical images, such as prostate cancer (PCa) lesions in PSMA PET images in certain embodiments.") or (b)radiomic features, pathomic features, or a combination of the radiomic features and the pathomic features (Fig. 2, para [056] "In some embodiments, the classification comprises outputting a predicted likelihood that the lesion is in a given prostate-specific membrane antigen reporting and data system (PSMA-RADS) class. In some embodiments, multiple predicted likelihoods are generated using an ensemble of CNNs and those predicted likelihoods are averaged to generate the classification.") extracted from the medical images of the patient, at least some of the medical images of the patient including at least one lesion, and at least some of the radiomic features or the pathomic features extracted from regions of the medical images of the patient including or around the at least one lesion (Fig. 2 para [053] "The present disclosure provides a deep learning (DL) framework to classify medical images, such as prostate cancer (PCa) lesions in PSMA PET images in certain embodiments.").
Claim 3: Du discloses the machine learning model of claim 1, wherein the machine learning model comprises a classifier (Fig. 2 para [053] "The present disclosure provides a deep learning (DL) framework to classify medical images, such as prostate cancer (PCa) lesions in PSMA PET images in certain embodiments.").
Claim 4, Du discloses the machine learning model of claim 1, wherein the one or more medical predictions comprise one or more of:
a diagnosis of a cancer (Figs. 15-17; para [0127] "The framework provided a confidence score for each prediction, which may help radiologists further interpret the output of the framework to make a more informed clinical diagnosis ... ", page 9 paragraph 0029 illustrating cancer);
a classification of the cancer according to phenotype or genotype;
a prognosis or prediction of cancer progression (Fig. 2, para [055] "Method of treating diseases, such as prostate cancer in subjects that utilize the classifications generated by these methods are also provided.");
a prediction of whether a particular lesion is likely to respond to a particular treatment (Fig. 2, para [055] "Method of treating diseases, such as prostate cancer in subjects that utilize the classifications generated by these methods are also provided.");
a prediction of whether apparent growth of a lesion represents true progression or a pseudo-progression (para [056] "In some embodiments, the classification comprises outputting a predicted likelihood that the lesion is in a given prostate-specific membrane antigen reporting and data system (PSMA-RADS) class."); or
a prediction of whether a particular patient is likely to experience a particular side effect.
Claim 5: Du discloses a method, comprising:
defining a loss function (Fig. 3B, para [075] "The tissue-type CNN classifier was trained on the training set separately from the rest of the framework on the tissue type classification task. The tissue-type CNN was trained on a per-slice basis by optimizing a class- weighted categorical cross-entropy loss function with an adaptive stochastic gradient descent-based optimization algorithm, Adam, using a batch size of 512 samples for 500 epochs.") that includes at least one negative predictive value (NPV) term (Figs. 8-14; para [0149] "Overall accuracy was computed across examples from all classes. Precision is defined as the number of true positives over the number of true and false positives. Recall is defined as the number of true positives over true positives and false negatives.", para [083] "The lesion-level performance of the trained framework was evaluated on both the validation and test sets. Evaluation metrics including overall accuracy, precision, recall, Fl score, receiver operating characteristic (ROC) curve, and area under ROC curve (AUROC) were assessed. Accuracy metrics, confusion matrices, and ROC curves were reported on a per-class basis and across all PSMA-RADS categories."); and
using the loss function, training a machine learning model to make one or more medical oncological predictions based on (a) one or more medical images of a patient (Fig. 2 para [053] "The present disclosure provides a deep learning (DL) framework to classify medical images, such as prostate cancer (PCa) lesions in PSMA PET images in certain embodiments."), or (b) radiomic features, pathomic features, or a combination of radiomic features and pathomic features extracted from the one or more medical images of the patient (Fig. 2, para [056] "In some embodiments, the classification comprises outputting a predicted likelihood that the lesion is in a given prostate-specific membrane antigen reporting and data system (PSMA-RADS) class. In some embodiments, multiple predicted likelihoods are generated using an ensemble of CNN s and those predicted likelihoods are averaged to generate the classification.", Fig. 2 para [053] "The present disclosure provides a deep learning (DL) framework to classify medical images, such as prostate cancer (PCa) lesions in PSMA PET images in certain embodiments.")).
Claim 8: Du discloses the method of claim 5, wherein the loss function further includes at least one overall model accuracy term (Figs. 8-14; para [0149] "Overall accuracy was computed across examples from all classes. Precision is defined as the number of true positives over the number of true and false positives. Recall is defined as the number of true positives over true positives and false negatives.").
Claim 9: Du discloses the method of claim 8, wherein the overall model accuracy term comprises a binary cross entropy (BCE) term or a positive predictive value (PPV) term (Fig. 3B, para [075] "The tissue-type CNN was trained on a per-slice basis by optimizing a class- weighted categorical cross-entropy loss function with an adaptive stochastic gradient descent-based optimization algorithm, Adam, using a batch size of 512 samples for 500 epochs. Evaluation metrics including overall accuracy, precision, recall, Fl score, receiver operating characteristic (ROC) curve, and area under ROC curve (AUROC) were assessed on the validation and test sets.").
Claim 10: Du discloses the method of claim 8, wherein the at least one at least one term and the overall model accuracy term are weighted (Fig. 3B, para [075] "The tissue-type CNN was trained on a per-slice basis by optimizing a class- weighted categorical cross-entropy loss function with an adaptive stochastic gradient descent-based optimization algorithm, Adam, using a batch size of 512 samples for 500 epochs.").
Claim 11: Du discloses the method of claim 5, wherein the loss function is continuously differentiable (para [078]).
Claim(s) 12 recite(s) substantially similar limitations as those of claim(s) 1 above, and are therefore rejected for substantially similar rationale as applied above, and incorporated herein.
Claim 14: Du discloses the method of claim 12, further comprising:
prior to said providing, extracting the set of features from the medical images (Fig. 1, para [054] "Method 100 also includes extracting one or more radiomic features from the PET and/or CT image to generate radiomic feature data (step 104) and combining the CNN-extracted image feature data and the radiomic feature data with anatomical location information about the lesion to generate combined information (step 106).").
Claim 15: Du discloses the method of claim 14, wherein said extracting comprises:
constructing a three-dimensional segmentation of one or more of: a lesion shown in the medical images, a peri-lesional region, or vasculature associated with the lesion (Fig. 1, para [054] "Method 100 also includes extracting one or more radiomic features from the PET and/or CT image to generate radiomic feature data (step 104) and combining the CNN-extracted image feature data and the radiomic feature data with anatomical location information about the lesion to generate combined information (step 106). In addition, method 100 also includes inputting the combined information into an artificial neural network (ANN) that classifies the lesion in the PET and/or CT image using the combined information to generate a classification (step 108)."); and
extracting at least some of the set of features from the three-dimensional segmentation (para [088]).
Claim 16: Du discloses the method of claim 12, wherein the loss function further includes at least one overall model accuracy term (Figs. 8-14; para [0149] "Overall accuracy was computed across examples from all classes. Precision is defined as the number of true positives over the number of true and false positives. Recall is defined as the number of true positives over true positives and false negatives.").
Claim 17: Du discloses the method of claim 16, wherein the overall model accuracy term
comprises a binary cross entropy (BCE) term or a positive predictive value (PPV) term (Fig. 3B, para [075] "The tissue-type CNN was trained on a per-slice basis by optimizing a class- weighted categorical cross-entropy loss function with an adaptive stochastic gradient descent-based optimization algorithm, Adam, using a batch size of 512 samples for 500 epochs. Evaluation metrics including overall accuracy, precision, recall, Fl score, receiver operating characteristic (ROC) curve, and area under ROC curve (AUROC) were assessed on the validation and test sets.").
Claim 18: Du discloses the method of claim 17, wherein the at least one term and the overall model accuracy term are weighted (Fig. 3B, para [075] "The tissue-type CNN was trained on a per-slice basis by optimizing a class- weighted categorical cross-entropy loss function with an adaptive stochastic gradient descent-based optimization algorithm, Adam, using a batch size of 512 samples for 500 epochs.").
Claim 19: Du discloses the method of claim 12, wherein the loss function is continuously
differentiable (para [078]).
Claim 22: Du discloses a machine-readable medium containing a set of machine- readable instructions that, when executed by a machine, cause the machine to perform the method of claim 12 (para [020]).
Claim(s) 23, 24 recite(s) substantially similar limitations as those of claim(s) 1, 1 above, and are therefore rejected for substantially similar rationale as applied above, and incorporated herein.
Claim 25: Du discloses the system of claim 24, wherein the machine learning model is pretrained to make the medical prediction using features extracted from the one or more medical images (para [065] "A cropped PET image slice containing a lesion, radiomic features, and anatomical information extracted from that lesion were used as inputs (FIG. 3). A deep convolutional neural network (CNN) extracted lesion features directly from the cropped PET image slice. The CNN implicitly extracted textural information and local contextual features in early layers of the network as well as global information in later layers relevant for the classification task").
Claim 27: Du discloses the system of claim 25, further comprising: at least one feature extraction module adapted to extract the features from the medical images (Fig. 1, para [054] "Method 100 also includes extracting one or more radiomic features from the PET and/or CT image to generate radiomic feature data (step 104) and combining the CNN-extracted image feature data and the radiomic feature data with anatomical location information about the lesion to generate combined information (step 106).").
Claim(s) 30 recite(s) substantially similar limitations as those of claim(s) 15 above, and are therefore rejected for substantially similar rationale as applied above, and incorporated herein.
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.
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.
Claim(s) 31-32 is/are rejected under 35 U.S.C. 103 as being unpatentable over Du in view of Cahill (20070036402).
Claim 31: Du discloses:
The machine learning model of claim 30, as discussed above and incorporated herein.
Du does not disclose:
wherein the features extracted from the vasculature are descriptive of one or both of the tortuosity or curvature of the vasculature.
Cahill discloses:
wherein the features extracted from the vasculature are descriptive of one or both of the tortuosity or curvature of the vasculature (page 7 paragraph 0082 illustrating determining curvatures in a patient’s image).
Before the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art to include the curvature detection of Cahill within the system of Du with the motivation of improving patient care by leveraging known image processing techniques (page 2 paragraph 0022).
Claim(s) 32 recite(s) substantially similar limitations as those of claim(s) 31 above, and are therefore rejected for substantially similar rationale as applied above, and incorporated herein.
Response to Arguments
In the Remarks filed on 06 February 2026, Applicant makes numerous arguments. Examiner will address these arguments in the order presented.
On page 6-7 Applicant argues that claim 22 is directed towards at least one statutory category of invention.
In making this argument, neither the Specification as originally filed nor the arguments provide a controlling definition for the machine-readable medium.
The BRI of machine readable media can encompass non-statutory transitory forms of signal transmission, such as a propagating electrical or electromagnetic signal per se. MPEP 2016.03(II)
On page 7-8 Applicant argues that the claims are statutory by being a statutory category other than a process.
Because both product and process claims may recite a "mental process", the phrase "mental processes" should be understood as referring to the type of abstract idea, and not to the statutory category of the claim. The courts have identified numerous product claims as reciting mental process-type abstract ideas, for instance the product claims to computer systems and computer-readable media in Versata Dev. Group. v. SAP Am., Inc., 793 F.3d 1306, 115 USPQ2d 1681 (Fed. Cir. 2015). MPEP 2106.04(a)(2)(III)
On page 9-12 Applicant argues that the claims provide technical improvement.
While Applicant’s arguments have been carefully considered, they are not found persuasive because the arguments are directed towards features that are part of the abstract idea.
Even newly discovered or novel judicial exceptions are still exceptions. MPEP 2106.04(I)
The additional elements do not provide any technical improvements because they merely invoke the machine learning in a generic manner.
Specifically, on page 11-12 Applicant contends that the claims cannot be practically performed in the human mind.
In making this argument, Applicant provides no evidence. Examiner maintains that the data processing as recited may be practically performed in the human mind for the reasons stated in the section above, and incorporated herein.
Additionally, the claims do not improve machine learning because the argued improvements are, at best, part of the abstract idea.
On page 12-13 Applicant argues that Du does not disclose cross-entropy loss functions.
In response to applicant's argument that the references fail to show certain features of the invention, it is noted that the features upon which applicant relies (i.e., cross-entropy loss functions; portions of Specification cited by Applicant discloses “for example”, and are considered to be mere non-committal exemplary embodiments, and not controlling definitions) are not recited in the rejected claim(s). Although the claims are interpreted in light of the specification, limitations from the specification are not read into the claims. See In re Van Geuns, 988 F.2d 1181, 26 USPQ2d 1057 (Fed. Cir. 1993).
Similar rationale applies to Applicant’s arguments on page 13 with respect to claim 23.
Based on the evidence presented above, Applicant’s arguments are not found persuasive.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Rabinowitz (20070027636) discloses a mathematical used to track phenotypes of a patient (Abstract) in a manner similar to those disclosed in the instant pending Specification as originally filed.
Wickert (20140278547) discloses predicting patient outcome using a patient’s medical history (Abstract) in a manner similar to those disclosed in the instant pending Specification as originally filed.
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 extension fee 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 date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to TRAN N NGUYEN whose telephone number is (571)272-0259. The examiner can normally be reached Monday-Friday 9AM-5PM Eastern.
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, KAMBIZ ABDI can be reached on (571)272-6702. 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.
/T.N.N./ Examiner, Art Unit 3685
/KAMBIZ ABDI/Supervisory Patent Examiner, Art Unit 3685