DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Status of the Claims
The office action is in response to the claims filed on February 17, 2025 for the application filed February 17, 2025 which claims priority to a foreign application filed on March 17, 2023. Claims 1-15 are currently pending and have been examined.
Claim Objections
Claims 9-11 and 13 are objected to because of the following informalities: The first recitations of the acronyms IGF and AI in claims 9-11 and 13 should provide the definition of the acronyms, similar to claim 8. Appropriate correction is required.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 6, 9-10 and 12 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 6 recites the limitation " wherein providing, for display, the explanations comprises ". There is insufficient antecedent basis for limitation in the claim as claim 1 does not recite “for providing, for display…”.
Claim 9 recites the limitations “the Graph AI” and “the input data”. There is insufficient antecedent basis for these limitations in the claim.
Claim 12 recites the limitations “the Kullback-Leibler (KL) divergence” and “the cosine similarity function”. There is insufficient antecedent basis for these limitations in the claim.
Claim 10 is rejected based on its dependency on claim 9.
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 1-15 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more.
Eligibility Step 1:
Under step 1 of the 2019 Revised Patent Subject Matter Eligibility Guidance, claims 1-13 are directed towards a computer-implemented method (i.e. a process), which is a statutory category. Claim 14 is directed towards a computer system (i.e. a machine), which is a statutory category. Claim 15 is directed towards a tangible, non-transitory computer-readable medium (i.e. a manufacture), which is a statutory category. Since the claims are directed toward statutory categories, it must be determined if the claims are directed towards a judicial exception (i.e. a law of nature, a natural phenomenon, or an abstract idea). In the instant application, the claims are directed towards an abstract idea.
Eligibility Step 2A, Prong One:
Under step 2A, prong one of the 2019 Revised Patent Subject Matter Eligibility Guidance, independent claims 1, 14 and 15 are determined to be directed to an judicial exception because an abstract idea is recited in the claims which fall within the subject matter groupings of abstract ideas. The abstract idea (identified in bold) recited in the representative claim 1 is identified as:
A computer-implemented method for improving explainability of patient representations, comprising:
generating one or more patient representations of a patient based on building one or more invariant feature representations of the patient, wherein the one or more patient representations indicate one or more discrete features;
determining predictions for one or more downstream tasks based on using the one or more discrete features; and
providing explanations associated with the one or more discrete features, wherein the explanations are associated with the predictions for the one or more downstream tasks.
The identified limitations fall within the subject matter grouping of certain methods of organizing human activity related and the sub grouping of managing personal behavior or relationships or interactions between people, (including social activities, teaching, and following rules or instructions). The claims recite a method of irganziing the human actiivyt of generating patient representations, determining predictions and providing explanations of the predictions, which are activities routinely performed by healthcare professionals when providing predictions and explanations based on patient data/representations and therefore a method organizing personal behavior of a healthcare professional.
The identified limitations also fall within the subject matter grouping of mental processes. The claimed generating patient representations, determining predictions and providing explanations can be performed in the human mind with the aid of pen and paper using observations, evaluations, judgments and opinions. If a claim recites a limitation that can practically be performed in the human mind, with or without the use of a physical aid such as pen and paper, the limitation falls within the mental processes grouping, and the claim recites an abstract idea.
Accordingly, claims 1, 14 and 15 recite an abstract idea under step 2A, prong one.
Eligibility Step 2A, Prong Two:
Under step 2A, prong two of the 2019 Revised Patent Subject Matter Eligibility Guidance, it must be determined whether the identified abstract ideas are integrated into a practical application. After evaluation, there is no indication that any additional elements or combination of elements integrate the abstract idea into a practical application, such as through: an additional element that reflects an improvement to the functioning of a computer, or an improvements to any other technology or technical field; an additional element that applies or uses a judicial exception to effect a particular treatment or prophylaxis for a disease or medical condition; an additional element that implements the judicial exception with, or uses the judicial exception in connection with, a particular machine or manufacture that is integral to the claim; an additional element that effects a transformation or reduction of a particular article to a different state or thing; or an additional element that applies or uses the judicial exception in some other meaningful way beyond generally linking the use of the judicial exception to a particular technological environment, such that the claim as a whole is more than a drafting effort designed to monopolize the exception. As shown below, the additional elements, other than the abstract idea per se, when considered both individually and as an ordered combination, amount to no more than a recitation of: generally linking the abstract idea to a particular technological environment or field of use; insignificant extra-solution activity to the judicial exception; and/or adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely using a computer as a tool to perform an abstract idea as evidenced below.
The additional elements recited in representative claim 1 are identified in italics as:
A computer-implemented method for improving explainability of patient representations, comprising:
generating one or more patient representations of a patient based on building one or more invariant feature representations of the patient, wherein the one or more patient representations indicate one or more discrete features;
determining predictions for one or more downstream tasks based on using the one or more discrete features; and
providing explanations associated with the one or more discrete features, wherein the explanations are associated with the predictions for the one or more downstream tasks.
The additional limitations of “computer-implemented” of claim 1, “A computer system, the system comprising one or more hardware processors” of claim 14 and “a tangible, non-transitory computer-readable medium having instructions thereon which, upon being executed by one or more processors” of claim 15 are determined to be mere instructions to apply an abstract idea under MPEP §2106.05(f). These limitations are recited at a high level of generality and merely used to implement the identified abstract idea. Therefore, these additional elements amount to no more than a recitation of the words "apply it" (or an equivalent) or no more than mere instructions to implement an abstract idea or other exception on a computer or no more than merely using a computer as a tool to perform an abstract idea.
Accordingly, claims 1, 14 and 15 do not recite additional elements which integrate the abstract idea into a practical application.
Eligibility Step 2B:
Under step 2B of the 2019 Revised Patent Subject Matter Eligibility Guidance, it must be determined whether provide an inventive concept by determining if the claims include additional elements or a combination of elements that are sufficient to amount to significantly more than the judicial exception. After evaluation, there is no indication that an additional element or combination of elements 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 limitations are determined to be mere instructions to apply an abstract idea under MPEP §2106.05(f), which do not amount to significantly more than the abstract idea.
Furthermore, 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 amounts to an inventive concept.
Dependent Claims:
The dependent claims merely present additional abstract information in tandem with further details regarding the elements from the independent claims and are, therefore, directed to an abstract idea for similar reasons as given above. None of these limitations are deemed to integrate the claims into a practical application or to amount to significantly more than the abstract idea as detailed below.
Regarding claim 2, the collecting of data to create a EHR used to generate the patient representations is determined to be no more than insignificant extra-solution activity to the judicial exception of mere data gathering under MPEP §2106.05(g), which is well-understood, routine and conventional as evidenced by MPEP §2106.05(d), subsection II
Regarding claim 3, the generating of patient biomarkers as part of the patient representation is considered to be encompassed by the abstract idea groupings identified with respect to claim 1.
Regarding claim 4, training a model to determine the predictions is determined to be mere instructions to apply an abstract idea using machine learning under MPEP §2106.05(f) as the claim does not provide any details as to how the training or determining is accomplished.
Regarding claim 5, the predicting and detecting are considered to be encompassed by the abstract idea groupings identified with respect to claim 1. The use of the trained model is determined to be mere instructions to apply an abstract idea using machine learning under MPEP §2106.05(f).
Regarding claim 6, the use of a device for the providing is determined to be mere instructions to apply an abstract idea using machine learning under MPEP §2106.05(f).
Regarding claims 7-13, the determining of invariant features are considered to be encompassed by the abstract idea groupings identified with respect to claim 1. The use of graph AI is determined to be mere instructions to apply an abstract idea using machine learning under MPEP §2106.05(f). The associations with loss functions and KNN algorithms are determined to fall within the subject matter grouping of mathematical concepts and/or to be no more than generally linking the use of a judicial exception to a particular technological environment or field of use under MPEP §2106.05(h).
Therefore, whether taken individually or as an ordered combination, 1-15 are rejected under 35 U.S.C. 101 as being directed to non-statutory subject matter.
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 set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied 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 1-15 are rejected under 35 U.S.C. 103 as being unpatentable over Malone et al. (U.S. Pub. No. 2019/0325995) in view of Dalli et al. (U.S. Pub. No. 2022/0114417).
Claim 1 (Original): A computer-implemented method for improving explainability of patient representations, comprising:
generating one or more patient representations of a patient based on building one or more invariant feature representations of the patient, wherein the one or more patient representations indicate one or more discrete features (Abstract, receiving a current episode snapshot of the caretaker episode comprising multi-modal data of the patient from an electronic health records (EHR) system, the multi-modal data including one or more available data modalities and one or more missing data modalities. The multi-modal data is applied as input to an embedding model having a submodel for each of the data modalities. A first embedding is generated for each of the available data modalities. A second embedding is generated for each of the missing data modalities using corresponding embeddings of neighbors in an episode snapshot graph. The first and second embeddings are combined to obtain a complete embedding. Paragraph [0092], . Each submodel is a function with a set of parameters which takes as input the respective data modality and outputs dense numeric vectors as the common embedding. Also see paragraph [0123].);
determining predictions for one or more downstream tasks based on using the one or more discrete features (Abstract, The patient outcome is predicted based on the complete embedding for the current episode snapshot using a machine learning component which has been trained using patient outcomes of the historical episode snapshots. Paragraph [0109], This complete embedding is used in a step S14 as input to the trained machine learning model of the machine learning component 18, which predicts the patient outcome. Then, in step S15, the predicted patient outcome including, for example, LOS and DD is used to plan hospital and healthcare resources, for scheduling purposes and/or for alerting respective DDs in advance of a predicted discharge to that location at a predicated time. Paragraph [0115], If desired, the embeddings can then be used in a traditional downstream machine learning task. In the context of EHRs, patients are modeled as nodes in the graph and similarity relationships between patients are modeled with edges. Paragraph [0116], Combining these learned feature embeddings gives a complete embedding for downstream tasks. Also see paragraph [0123].); and
providing explanations associated with the one or more discrete features, wherein the explanations are associated with the predictions for the one or more downstream tasks (Paragraph [0115], since EP learns in a first step embeddings for each data modality independently, the resulting predictions are more interpretable as it is possible to distinguish between the influence of these independently learned modality embeddings on the predictions. For instance, it is possible to assess the influence of the free text data and the time series data.).
Malone further discloses providing interpretable predictions associated with the one or more discrete features, wherein the interpretable predictions are associated with the predictions for the one or more downstream tasks (Paragraph [0115], since EP learns in a first step embeddings for each data modality independently, the resulting predictions are more interpretable as it is possible to distinguish between the influence of these independently learned modality embeddings on the predictions. For instance, it is possible to assess the influence of the free text data and the time series data. Paragraph [0115], If desired, the embeddings can then be used in a traditional downstream machine learning task. In the context of EHRs, patients are modeled as nodes in the graph and similarity relationships between patients are modeled with edges. Paragraph [0116], Combining these learned feature embeddings gives a complete embedding for downstream tasks. Also see paragraph [0123].), but does not appear to explicitly disclose that the interpretable predictions include explanations.
Dalli teaches that it was old and well known in the art of artificial intelligence at the time of the filing to include providing explanations associated with the one or more discrete features, wherein the explanations are associated with the predictions for the one or more downstream tasks (Dalli, paragraph [0044], An interpretable system or model may be capable of generating a model explanation accompanying the answer output. Also see paragraph [0157] and Fig. 3.) to achieve better performance in meeting the user goals and tasks (Dalli, paragraph [0244]).
Therefore, it would have been obvious to one of ordinary skill in the art of artificial intelligence at the time of the filing to modify the interpretable predictions of Malone to include explanations, as taught by Dalli, in order to achieve better performance in meeting the user goals and tasks.
Regarding claim 2, Malone further discloses: collecting data from a plurality of patients from different subsystems within a hospital environment; and creating an electronic health record (EHR) database based on the collected data, wherein generating the one or more patient representations is based on using the EHR database (Abstract, receiving a current episode snapshot of the caretaker episode comprising multi-modal data of the patient from an electronic health records (EHR) system, the multi-modal data including one or more available data modalities and one or more missing data modalities. The multi-modal data is applied as input to an embedding model having a submodel for each of the data modalities. A first embedding is generated for each of the available data modalities. A second embedding is generated for each of the missing data modalities using corresponding embeddings of neighbors in an episode snapshot graph. The first and second embeddings are combined to obtain a complete embedding. The patient outcome is predicted based on the complete embedding for the current episode snapshot using a machine learning component which has been trained using patient outcomes of the historical episode snapshots. Paragraph [0115], In the context of EHRs, patients are modeled as nodes in the graph and similarity relationships between patients are modeled with edges. Also see Figs. 1 and 4.).
Regarding claim 3, Malone further discloses wherein generating the one or more patient representations of the patient comprises generating biomarkers for the patient, wherein determining the predictions for the one or more downstream tasks is based on the generated biomarkers (Paragraph [0099], there is one submodel for demographics, one submodel for heart rate time series, one submodel for blood oxygen level time series, etc. Also see paragraphs [0043]-[0087].).
Regarding claim 4, Malone further discloses: training a model based on the biomarkers for the patient, wherein determining the predictions is based on the trained model (Paragraph [0109], In a step S3, the episode snapshot graph created in step S2 and data modalities from the historical episode snapshots are used for unsupervised training of an embedding model, in particular submodels for each of the data modalities, for example, as discussed above with reference to FIG. 3. In a step S4, a machine learning model, such as a regression model, of the machine learning component 18 is trained based on the embeddings of the historical episode snapshots and their corresponding patient outcomes (where available). Paragraph [0100], learns embeddings 52 for combined heart rate and blood pressure data as a data modality.
Regarding claim 5, Malone further discloses: predicting one or more risks for the patient based on using the trained model; and detecting, based on the one or more risks, specific biomarkers from the generated biomarkers that cause each of the predictions, wherein the explanations indicate the predictions and the specific biomarkers that caused the predictions (Paragraph [0018], the predictions could be used to identify patients that need to be prioritized and receive more care. Paragraph [0122], prediction tasks were performed to predict: in-hospital mortality (mort), LOS and DD. Paragraph [0015], it is possible to distinguish between the influence of these independently learned modality embeddings on the predictions. For instance, it is possible to assess the influence of the free text data and the time series data. Paragraph [0099], there is one submodel for demographics, one submodel for heart rate time series, one submodel for blood oxygen level time series, etc. Abstract, The multi-modal data is applied as input to an embedding model having a submodel for each of the data modalities. A first embedding is generated for each of the available data modalities.).
Regarding claim 6, Malone further discloses wherein providing,Paragraph [0109], in step S15, the predicted patient outcome including, for example, LOS and DD is used to plan hospital and healthcare resources, for scheduling purposes and/or for alerting respective DDs in advance of a predicted discharge to that location at a predicated time. Also see paragraph [0018]. When viewed as a whole, the disclosure of Malone is implemented in a computer environment such that the interpretable predictions and explanations would be provided on a device. The limitation “for display’ in construed as intended used and not given patentable weight. Also see Dalli, paragraphs [0011] and [0031].).
Regarding claim 7, Malone further discloses wherein the one or more discrete features comprise invariant graph fingerprint (IGF) features, wherein the explanations are associated with the IGF features, and wherein the explanations indicate importance of the IGF featuresParagraph [0145], EP was used to learn embeddings for each of the four data modalities (time series features, text notes, demographics, and episode identity within the graph) independently. The final representation for the episode is the concatenation of all modality embeddings. Paragraph [0115], the resulting predictions are more interpretable as it is possible to distinguish between the influence of these independently learned modality embeddings on the predictions.).
Malone does not appear to explicitly disclose wherein the explanations indicating importance of the IGF features are according to Shapley importance explanations.
Dalli teaches that it was old and well known in the art of artificial intelligence at the time of the filing wherein the explanations indicate importance of the IGF features according to Shapley importance explanations (Dalli, paragraph [0161], An exemplary EIGS may also utilize various widgets and output formats to explain particular details in an explanation output, including but not limited to… Shapley values.) to achieve better performance in meeting the user goals and tasks (Dalli, paragraph [0244]).
Therefore, it would have been obvious to one of ordinary skill in the art of artificial intelligence at the time of the filing to modify the explanations of Malone to indicate importance of the IGF features according to Shapley importance explanations, as taught by Dalli, in order to achieve better performance in meeting the user goals and tasks.
Regarding claim 8, Malone further discloses wherein generating the one or more patient representations of the patient comprises determining a first invariant graph fingerprint (IGF) feature for input features based on using a graph artificial intelligence (Graph AI) and input data, wherein the first IGF feature is a discrete version of the input data (Paragraph [0109], In a step S3, the episode snapshot graph created in step S2 and data modalities from the historical episode snapshots are used for unsupervised training of an embedding model, in particular submodels for each of the data modalities, for example, as discussed above with reference to FIG. 3. In the prediction method, which is preferably performed online, a new or current episode snapshot is extracted, for example, from one or more EHR systems 21 in a step S10. For the available data modalities of the new episode snapshot, the respective trained submodels of the embedding model 16 can be used to generate a first embedding in each case in a step S12. For the missing data modalities, a second embedding in each case is generated using neighbors in the episode snapshot graph 40 in a step S11. For this purpose, the new episode snapshot is located in the episode snapshot graph based on the similarity measure defined from domain knowledge, and message passing can be used to generate the embeddings for the missing data modalities in accordance with any of the embodiments discussed herein. In a step S13, the first and second embeddings are concatenated to form a complete embedding for the new episode snapshot. Paragraph [0092], Each submodel is a function with a set of parameters which takes as input the respective data modality and outputs dense numeric vectors as the common embedding. Also see paragraph [0115].).
Regarding claim 9, Malone further discloses wherein generating the one or more patient representations of the patient comprises determining, based on using the Graph AI and the input data, a second IGF feature for prediction tasks and a third IGF feature for prototypes, wherein the second IGF feature is a discrete subset of the input data that is used for a prediction of a specific task, and wherein the third IGF feature indicates a clustering of the input data associated with similarities between the one or more patient representations (Paragraph [0109], In the prediction method, which is preferably performed online, a new or current episode snapshot is extracted, for example, from one or more EHR systems 21 in a step S10. For the available data modalities of the new episode snapshot, the respective trained submodels of the embedding model 16 can be used to generate a first embedding in each case in a step S12. For the missing data modalities, a second embedding in each case is generated using neighbors in the episode snapshot graph 40 in a step S11. For this purpose, the new episode snapshot is located in the episode snapshot graph based on the similarity measure defined from domain knowledge, and message passing can be used to generate the embeddings for the missing data modalities in accordance with any of the embodiments discussed herein. In a step S13, the first and second embeddings are concatenated to form a complete embedding for the new episode snapshot. Paragraph [0092], Each submodel is a function with a set of parameters which takes as input the respective data modality and outputs dense numeric vectors as the common embedding. Paragraph [0130], Intuitively, for each patient node ν and attribute type i, the vector representation reconstructed from the embeddings of patient nodes neighboring ν are learned to be more similar to the embedding of attribute type i for ν than to the embedding of attribute type i of a random patient node in the graph. Paragraph [0147], extract the relevant data modalities for the relevant prediction task. Also see paragraphs [0018], [0100], [0115] and [0140]. The limitation “for protypes” is construed as intended use and not given patentable weight.).
Regarding claim 10, Malone further discloses wherein determining the third IGF feature for prototypes is based on using one or more generated virtual nodes and adding features that are determined using a k-Nearest neighbor algorithm (Paragraph [0109], For the missing data modalities, a second embedding in each case is generated using neighbors in the episode snapshot graph 40 in a step S11. For this purpose, the new episode snapshot is located in the episode snapshot graph based on the similarity measure defined from domain knowledge, and message passing can be used to generate the embeddings for the missing data modalities in accordance with any of the embodiments discussed herein. In a step S13, the first and second embeddings are concatenated to form a complete embedding for the new episode snapshot. Paragraph [0130], Intuitively, for each patient node ν and attribute type i, the vector representation reconstructed from the embeddings of patient nodes neighboring ν are learned to be more similar to the embedding of attribute type i for ν than to the embedding of attribute type i of a random patient node in the graph. Paragraph [0140], The similarity between two episodes is then defined as i and j as: s i,j=exp−d i,j (Equation 6) where di,j is the Euclidean distance between the respective sentence embeddings. Finally, all pairs of episodes are connected for which si,j>0.9. Also see paragraph [0115]. The limitation “for protypes” is construed as intended use and not given patentable weight.).
Regarding claim 11, Malone further discloses wherein generating the one or more patient representations of the patient comprises determining a fourth IGF feature for counterfactuals and determining a fifth IGF feature for a contrastive associated with a contrastive loss (Paragraph [0100], submodel 50 consists of two function, fd and fm. The first function (fd) embeds all observed raw patient data 46 (indicated by the filled-in circles of different shades), while the second function (fm) gives an embedding for missing raw patient data 46 (indicated by “x”) for the episode snapshot 42 a. Paragraph [0133], During training, EP learns how to reconstruct the missing data representation based on a contrastive loss between representations of existing labels, or embeddings of episode snapshots (see FIG. 3), and, therefore, can learn how to reconstruct a representation when data is missing. Also see paragraphs [0101] and [0134]. The limitations “for counterfactuals” and “for a contrastive” are construed as intended use and not given patentable weight.).
Regarding claim 12, Malone further discloses wherein the contrastive loss is associated with Paragraphs [0101], [0133]-[0134] and [0140].).
Malone does not appear to explicitly disclose that the similarity function is a
Dalli teaches that it was old and well known in the art of artificial intelligence at the time of the filing that a similarity function may be a cosine similarity function (Dalli, paragraph [0239], Similarity metrics, such as cosine similarity, may be used ) to achieve better performance in meeting the user goals and tasks (Dalli, paragraph [0244]).
Therefore, it would have been obvious to one of ordinary skill in the art of artificial intelligence at the time of the filing to modify the similarity function of Malone to be a cosine similarity function, as taught by Dalli, in order to achieve better performance in meeting the user goals and tasks.
Regarding claim 13, Malone further discloses wherein generating the one or more patient representations of the patient comprises determining one or more IGF features based on using a dedicated loss or one or more unsupervised computations (Paragraph [0133], unsupervised loss function used in EP. Also see paragraph [0115].).
Regarding claims 14-15: all limitations as recited have been analyzed and rejected with respect to claim 1. Claims 14 pertains to a computer system, corresponding to the computer-implemented method of claim 1. Claims 15 pertains to a tangible, non-transitory computer-readable medium, corresponding to the computer-implemented method of claim 1. Claims 14-15 do not teach or define any new limitations beyond claims 1 apart from the processor, memory and medium, disclosed by Malone in paragraph [0030]; therefore claims 14-15 are rejected under the same rationale.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Devin C. Hein whose telephone number is (303)297-4305. The examiner can normally be reached 9:00 AM - 5:00 PM M-F MDT.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Jason B. Dunham can be reached at (571) 272-8109. 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.
/DEVIN C HEIN/Examiner, Art Unit 3686
Prosecution Insights