Detailed Action
Status of Claims
This action is in reply to applicant response filed on October 15, 2025.
Claims 1, 6-9, and 17 have been amended.
Claims 1-20 are currently pending and have been examined.
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 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1: Determining that a claim falls within one of the four enumerated categories of patentable subject matter recited in 35 U.S.C. 101 (i.e., process, machine, manufacture, or composition of matter). (MPEP 2106.03)
Claims 1-8 and 17-20 describe tangible system components, thus falling within one of the four statutory classes; i.e., machine or manufacture.
Claims 9-16 recite a series of steps, thus falling within one of the four statutory classes; i.e., a process.
Step 2A, Prong One: Identify Whether the Claims Are Directed to a Judicial Exception.
USPTO guidance identifies the following abstract idea categories:
1) Mathematical concepts; 2) Certain methods of organizing human activity; 3) Mental processes (including collecting, analyzing, and displaying information).
- Collecting information: Accessing cardiac data, searching publication repositories.
- Analyzing information: Filtering publications for recency and relevance.
- Reporting information: Generating a natural language report.
- Use of a neural network: Recited at a high level, with no specific technical improvement or specialized architecture.
These steps are similar to those found abstract in Federal Circuit cases (e.g., *Electric Power Group*; *Content Extraction & Transmission*; *Cleveland Clinic*).
As such, the claims recite an abstract idea: collecting, analyzing, and reporting information, which is a mental process and a fundamental practice long prevalent in the field.
Step 2A, Prong Two: Does the Claim Recite Additional Elements That Integrate the Exception Into a Practical Application?
- The claims recite implementation on a generic processor and memory.
- The neural network is not claimed with any specific, unconventional architecture or technical improvement.
- The publication repository is a generic data source (e.g., a website).
- No improvement to the functioning of a computer or another technology is recited.
- The result (a natural language report) is merely displayed or rendered, not used to control a physical process or effect a transformation.
The claims, considered as a whole, do not integrate the abstract idea into a practical application. They amount to automating a conventional workflow using generic computer technology.
Relevant Precedent:
- *Electric Power Group*: Merely collecting, analyzing, and displaying information is not a practical application.
- *SAP America v. InvestPic*: Even advanced mathematical models are not a practical application if used for information analysis and reporting.
Step 2B: Do the Claims Recite Significantly More Than the Abstract Idea?- The claims recite generic computer components (processor, memory).
- The steps of accessing data, filtering, searching, and generating a report are all routine and conventional in the art.
- The use of a neural network is generic; there are no details of a novel training technique, architecture, or other technical improvement.
- The claims do not recite a transformation of a physical article or a particular machine.
- The combination of steps does not yield an unconventional or non-generic arrangement.
The claims do not recite any additional element or combination of elements that amounts to significantly more than the abstract idea.
Cited Cases:
- *Alice Corp. v. CLS Bank*: Generic computer implementation of an abstract idea is not significantly more.
- *In re TLI Communications*: Generic data storage and analysis on a computer is not an inventive concept.
- *Cleveland Clinic*: Using a known machine learning technique for data analysis/reporting is not inventive.
Therefore, the claims are directed to an abstract idea, namely, collecting, analyzing, and reporting information (i.e., accessing patient data, filtering and searching publication repositories, and generating a report). The claims do not recite any additional elements that integrate the abstract idea into a practical application. The steps are performed using generic computer components (processor, memory), and the use of a neural network is recited at a high level without any technical detail or improvement to computer technology. The claims elements, individually and in combination, do not amount to significantly more than the abstract idea. Therefore, the claim is not patent eligible under § 101.
Dependent claims: However, similar to the analysis of the additional elements identified in claims 1, 9, and 17 described above, the aforementioned additional identified for dependent claims 2-8, 10-16, and 18-20 also deemed to be: (1) no more than a recitation of: adding the words "apply it" (or an equivalent) with the judicial exception; mere instructions to implement an abstract idea on a computer; or merely uses a computer as a tool to perform an abstract idea; and (2) the equivalent of adding insignificant extra-solution activity to the judicial exception. See MPEP §§ § 2106.05(f), (g). Thus, the additional elements in dependent claims 2-8, 10-16, and 18-20 are not indicative of integrating the judicial exception into a practical application.
Therefore, claims 1-20 are not patent eligible.
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 of this title, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1, 4-9, 12-17, and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Yu, et al. (“Zero-Shot ECG Diagnosis with Large Language Models and Retrieval-Augmented Generation”, hereinafter Yu) in view of Eickhoff et al. (US 2020/0105419), hereinafter, Eickhoff).
As per claims 1, 9, and 17, Yu discloses a computer-implemented method, comprising:
accessing, by a device operatively coupled to a processor, cardiac data associated with a medical patient (3.1 “clinical 12-lead ECG records”);
searching, by the device (4.1 “This setup facilitates the search and retrieval of related text from the embedding space with appropriate prompts.”, Appendix A. “consider the supplemental information from textbooks regarding the detected features.” Figure 1 “ECG-specific knowledge documents”); and
generating, by the device and by executing the deep learning neural network on both the cardiac data and the one or more cardiology publications, a natural language cardiology report for the medical patient (Figure 1, Step 2: “Zero-shot Diagnosis” and LLM Structured Output. See also Appendix A describing an example application).
Yu does not appear to explicitly teach filtering out from a dynamic cardiology publication repository each cardiology publication whose publication date is before that on which a deep learning neural network was most recently trained and searching after the filtering (
However, Eickhoff discloses an automated, literature-driven disease diagnosis support system that processes patient data, retrieves and analyzes relevant medical literature, and suggests diagnoses based on fused and filtered concepts, with user-customizable query preprocessing and filtering options (abstract and claims).
Eickhoff further teaches a preprocessing operation may be performed on each of the queries. The preprocessing operation may be performed in order to filter out keywords in a query that do not add value to the diagnosis prediction process and to remove an uninformative keywords from the one or more keywords. For example, from the content of a query, keywords such as stop words, uninformative e part-of-speech tags such as verbs, determiners, adpositions, coordinating conjunctions, and punctuations can be removed. The system can customize the type of keywords to include and the type of keywords to exclude as part of the query preprocessing operation, such that a user may have the option to customize the query preprocessing operation. Therefore, it would be an obvious matter of design choice for a user to customize the query preprocessing operation by including a date range filter in order to retrieve the most recent publications. ¶ 42 & fig.3 RC 312).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the retrieval-augmented neural network diagnosis/reporting system of Yu with the customizable query and document filtering system of Eickhoff. Eickhoff teaches that queries can be customized to filter documents based on user-selected criteria, such as keywords, semantic types, or other attributes (see Eickhoff, ¶¶ 36, 37, 42, Fig. 3, "customize the type of keywords to include and the type of keywords to exclude as part of the query preprocessing operation").
While Eickhoff does not explicitly disclose a date range filter, it would have been an obvious matter of design choice to include a date range as a document attribute for filtering, as publication date is a common and well-understood attribute in information retrieval systems.
In particular, in the context of medical literature retrieval for automated diagnosis/reporting, a person of ordinary skill would recognize that filtering documents by publication date is a routine and predictable way to ensure the relevance and currency of retrieved information. Incorporating a date range filter as one of the customizable query options would have been straightforward and within the routine skill of the art, especially given Eickhoff's explicit teaching of customizable, user-defined query filters. This modification would not require more than ordinary skill and would predictably improve the system by allowing users to retrieve only the most recent or otherwise temporally relevant publications, addressing the recognized need for up-to-date clinical recommendations in automated reporting systems.
Therefore, the inclusion of a date range filter in Eickhoff's customizable query system, and its use in combination with Yu's retrieval-augmented neural network, would have been an obvious variation and matter of design choice to a person of ordinary skill in the art.
As per claims 4 and 12, the computer-implemented method of claim 9, Yu further discloses wherein the cardiac data comprises: an electrocardiogram of the medical patient; a photoplethysmogram of the medical patient; a seismocardiogram of the medical patient; or a phonocardiogram of the medical patient (3.1 “clinical 12-lead ECG records”).
As per claims 5 and 13, the computer-implemented method of claim 9, Yu further discloses wherein the cardiac data comprises physical characteristics or attributes of the medical patient (3.1 “clinical 12-lead ECG records”).
As per claims 6 and 14, the computer-implemented method of claim 9, Yu further discloses wherein the device searches the dynamic cardiology publication repository via: a keyword search technique; an embedding search technique; or a probabilistic information retrieval search technique (4.1. Construct Database of Domain Knowledge. “…search and retrieval of related text from the embedding space with appropriate prompts.”)
As per claims 7 and 15, the computer-implemented method of claim 9, Yu further discloses further comprising: visually rendering, by the device, the natural language cardiology report on an electronic display (Figure 1, Structured Output).
As per claims 8 and 16, the computer-implemented method of claim 15, Yu further discloses further comprising: prompting, by the device and on the electronic display, a user to accept or reject the natural language cardiology report (5.2 Evaluation Results).
As per claim 19, the computer program product of claim 17, Yu further discloses wherein the medical data comprises: a body temperature of the medical patient; a pulse rate of the medical patient; a respiration rate of the medical patient; or a blood pressure of the medical patient (4.2 heart rate).
Claims 2, 3, 10, 11, and 18 are rejected under 35 U.S.C. 103 as being unpatentable over Yu in view of Eickhoff, and further in view of Zakka et al. “Almanac: Retrieval-Augmented Language Models for Clinical Medicine” hereinafter, Zakka).
As per claims 2, 10, and 18 the computer-implemented method of claim 9, Yu does not explicitly disclose wherein the dynamic cardiology publication repository is a medical website, and wherein the one or more cardiology publications are cardiology research papers or cardiology journal articles published by the medical website. However, Zakka teaches using “external tools (e.g. search engines, medical databases and calculators) to answer queries related to clinical concepts and latest treatment recommendations.”.
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to augment Yu’s external knowledge by implementing the external tools as taught by Zakka, in order to further refine Yu’s diagnostic methods, and since the claimed invention is merely a combination of old elements, and in the combination each element merely would have performed the same function as it did separately, and one of ordinary skill in the art would have recognized that the results of the combination were predictable.
As per claims 3 and 11, the computer-implemented method of claim 9, Yu does not explicitly disclose wherein the cardiac data comprises: one or more computed tomography images of a heart of the medical patient; one or more magnetic resonance images of the heart of the medical patient; one or more echocardiogram images of the heart of the medical patient; or one or more positron emission tomography images of the heart of the medical patient.
However, Yu’s method is applied in the realm of cardiology and includes e.g., ECG data and Zakka teaches medical data including an MRI (page 15).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to include MRI data as taught by Zakka in the Yu’s cardia data, in order to allow for additional diagnosis and since the claimed invention is merely a combination of old elements, and in the combination each element merely would have performed the same function as it did separately, and one of ordinary skill in the art would have recognized that the results of the combination were predictable.
Claim 20 is rejected under 35 U.S.C. 103 as being unpatentable over Yu in view of Eickhoff, and further in view of Marks et al. (US 2021/0391075), hereinafter, Marks).
As per claim 20, Yu does not explicitly disclose but Marks does teach wherein the medical data comprises a medical scanned image of the medical patient (¶ 32).
Marks discloses a system that analyzes comprehensive patient health data, including structured fields, free-text notes, genomic data, and medical images to automatically recommend and personalize relevant medical literature for healthcare professionals, using a combination of expert-system rules, AI techniques, and user feedback (abstract and claims).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the retrieval-augmented large language model (LLM) system of Yu et al., which utilizes patient physiological data (such as ECG) and relevant literature to generate diagnostic reports, with the teachings of Marks, which expressly discloses that the medical data analyzed by such a system may include medical scanned images of the patient (see Marks ¶ 32).
The motivation for this combination arises from the recognized need in the medical informatics and AI fields to provide more comprehensive and accurate diagnostic support by leveraging all available patient data types, including both physiological signals and medical images. Marks demonstrates that retrieval-augmented AI systems can be applied not only to textual or signal data, but also to medical scanned images, to improve the relevance and accuracy of diagnostic recommendations or reports.
A person of ordinary skill would have found it obvious to incorporate the use of medical scanned images as part of the input data in Yu’s system, as taught by Marks, in order to broaden the applicability and clinical utility of the retrieval-augmented AI system. Such a modification would have been a straightforward extension, requiring only routine adaptation, and would have predictably improved the system by enabling it to generate diagnostic reports based on a more complete set of patient information, including imaging data commonly used in clinical practice.
Therefore, the combination of Yu and Marks, resulting in a system in which the medical data may comprise medical scanned images of the patient, would have been obvious to a person of ordinary skill in the art.
Response To Arguments/Remarks
Applicant's amendments/remarks filed 10/15/2025 have been fully considered, as such the 112(a) & (b) rejections have been withdrawn. Also, In view of applicant’s amendments/remarks the previous rejection under 35 U.S.C. § 103 has been withdrawn. However, an updated search was conducted and an updated 35 U.S.C. § 103 has been applied.
Regarding 101: Thank you for your detailed remarks regarding the subject matter eligibility rejection under 35 U.S.C. § 101 for claims 1-20. The examiner has carefully considered applicant’s arguments and the cited portions of the specification. However, after review, the rejection is maintained for the reasons outlined below, consistent with the Alice/Mayo framework and current USPTO guidance.
The examiner’s position is that the claims are directed to the abstract idea of collecting, analyzing, and reporting information, specifically accessing patient cardiac data, filtering and searching publication repositories for relevant, recent documents, and generating a report using a neural network.
These steps constitute fundamental information processing, which falls within the categories of abstract ideas (mental processes and certain methods of organizing human activity), as recognized in *Electric Power Group v. Alstom S.A.*, *Content Extraction & Transmission LLC v. Wells Fargo Bank*, and the 2019 Revised Patent Subject Matter Eligibility Guidance.
The claims do not recite a particular improvement to computer technology, nor do they effect a transformation of a physical article. Rather, they automate a conventional workflow using generic computing components and a neural network applied in a routine manner.
While the applicant asserts that the claimed steps cannot be performed by the human mind (e.g., execution of a neural network), the use of a neural network alone does not confer eligibility, absent a specific technical improvement. The claims recite the use of a neural network at a high level, without details of a novel architecture, training method, or unconventional application. As established in *Cleveland Clinic Foundation v. True Health Diagnostics*, merely implementing a data analysis or reporting process via a neural network is not sufficient for eligibility.
The applicant argues the claims are not directed to organizing human activity. The examiner recognizes that the claims are not directed to human interactions per se, but rather to information processing. However, the automation of data gathering, filtering, and reporting, even when performed by a neural network, remains an abstract idea if not tied to a specific technical improvement or practical application beyond generic computer implementation.
The examiner does not categorically exclude generative AI inventions from eligibility. Eligibility depends on whether the claims recite a specific improvement to computer technology or a practical application of the abstract idea. Here, the claims do not recite such improvements; they recite the routine application of a neural network for synthesizing reports, which is not sufficient under *Alice* and subsequent guidance.
Applicant asserts that filtering publications based on their date relative to the last neural network training session is a technical improvement. However, this feature is a data selection step, which does not improve the functioning of the computer or neural network itself. It is a business rule for selecting relevant documents, not a technical solution to a computer-centric problem. The claims do not recite a novel mechanism for implementing this filtration, nor do they recite how this improves the underlying technology of neural networks or automated reporting systems. While the specification discusses reduced need for retraining, the claims do not recite a specific mechanism or technical solution that achieves this effect. The claimed filtration step is a conventional business logic for selecting data, not a technological improvement to the neural network or reporting process.
The cases cited by applicant (*Thales Visionix*, *Finjan*, *DDR Holdings*, *Enfish*, *Visual Memory*) are distinguishable:
- Thales Visionix: Claimed a specific improvement to inertial sensor fusion, not generic data analysis.
- Finjan: Claimed a specific mechanism for linking security profiles to downloadable content, not generic information analysis.
- DDR Holdings: Claimed a solution “rooted in computer technology” to address a problem unique to the Internet.
- Enfish/Visual Memory: Claimed specific improvements to database and computer memory architectures. In contrast, the instant claims do not recite a specific improvement to computer technology or a particular machine. The claimed steps are generic and abstract, lacking unconventional technical detail.
In conclusion, the claims do not integrate the abstract idea into a practical application. The steps are performed by generic processors and memory, and the neural network is applied in a routine manner. The output (a report) is rendered on a display, but this is a conventional result of information processing, not a transformation or improvement to computer technology. Further, the claims do not recite an inventive concept sufficient to confer eligibility. The use of a neural network and the training-date-based filtration are conventional steps, not unconventional or non-routine in the art. The claims do not recite a specific, technical improvement to the functioning of the computer or neural network.
For the reasons above, the examiner maintains the rejection under 35 U.S.C. § 101. The claims are directed to an abstract idea and do not integrate that idea into a practical application, nor do they recite an inventive concept.
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to FAHD A OBEID whose telephone number is (571) 270-3324. The examiner can normally be reached Monday-Friday 8:30am-5:00pm.
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.
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.
/FAHD A OBEID/ Supervisory Patent Examiner, Art Unit 3627
Prosecution Insights