Prosecution Insights
Last updated: April 19, 2026
Application No. 17/814,229

TRANSLATING AI ALGORITHMS FROM 12-LEAD CLINICAL ECGS TO PORTABLE AND CONSUMER ECGS WITH FEWER LEADS

Final Rejection §101
Filed
Jul 21, 2022
Examiner
MOSS, JAMES R
Art Unit
3792
Tech Center
3700 — Mechanical Engineering & Manufacturing
Assignee
Tempus AI Inc.
OA Round
6 (Final)
51%
Grant Probability
Moderate
7-8
OA Rounds
3y 3m
To Grant
92%
With Interview

Examiner Intelligence

Grants 51% of resolved cases
51%
Career Allow Rate
134 granted / 261 resolved
-18.7% vs TC avg
Strong +41% interview lift
Without
With
+41.0%
Interview Lift
resolved cases with interview
Typical timeline
3y 3m
Avg Prosecution
33 currently pending
Career history
294
Total Applications
across all art units

Statute-Specific Performance

§101
13.3%
-26.7% vs TC avg
§103
36.7%
-3.3% vs TC avg
§102
13.5%
-26.5% vs TC avg
§112
29.5%
-10.5% vs TC avg
Black line = Tech Center average estimate • Based on career data from 261 resolved cases

Office Action

§101
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 . Response to Arguments With regards to the 103 discussion Examiner disagrees with Applicant argument about Alb and Kwon regarding the generation of every lead. The paragraph applicants themselves recite that the reference discloses taking in the three leads and the “artificial” intelligence outputting/generating/provides the full 12 leads. Likewise, Kwon recites generating the 12-n in the first generative model and then based on the input of the 12-n leads generates the n leads in the second generative model see fig. 5 which is artificial. Regardless of Examiner disagreeing with the above arguments, the additional elements amended in and discussed in the arguments overcome the art. For clarity, Examiner notes that the specific new “selecting” elements appear to be supported by [0079] (using PG Pub for paragraph numbers) related to the claimed “first artificial intelligence model” (i.e. risk model taking in the artificial leads and determining a state) as opposed to the “second artificial intelligence model” (i.e. model which synthesizes additional leads from fewer input leads and outputs a full set of artificial leads). Applicant's arguments filed 8/21/25 have been fully considered but they are not persuasive. With regards to the 101 rejection Applicants argue it is eligible under 101, Examiner disagrees. Applicants first argue it cannot be performed in the mind. To the extent applicant is arguing things that are not claimed such as the frequency/volume of measurements etc. this is not persuasive. 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). Applicants point to the office’s guidance specifically Example 39, Examiner disagrees. Each claim is examined on the specifics of the facts in the particular case and while Example 39s claim was eligible, the claim is not analogous. Applicants’ claims do not recite the level of specifics in example 39. AI is a broad category that encompasses everything from majority vote to LLM’s. Examiner believes the mind is capable of performing the claimed functions and that the elements would fall into the observations, evaluations and/or judgements of the Kim memo. Applicants next argue that the Examiners rejection includes a false premise and that there is an “inherent level of accuracy” that is required, Examiner disagree. There is no discussion on this in the specification and without such a discussion how can the claims be read in view of it? Since there is no discussion in the specification Applicants argue it is “inherent”. With regards to Applicants argument with no recitation in the specification, what is the BRI or the “inherent level of accuracy”, 5%, 10%, 51%, 90%, 95%, indefinite etc.? Examiner notes that neither the model(s) nor the claim requires the inputs to be good just that there are inputs, thus for the BRI it could be any level of accuracy/inaccuracy. One could send imprecise/bad data in and get imprecise/bad output(s) to one or both of the models. With regards to the next part of this argument that there is no statement as to how one could perform this, there is no statement as to how one could perform this, as discussed in the previous rejection and response to arguments. To summarize, the human mind is capable of learning elements to build literal neural networks providing associations and make determinations based on inputs. For example, physicians have “trained” their minds to read ECGs and make determinations based on this data. Furthermore, a person is capable of training a model (such as linear regression etc.) mentally either in their head (or using pen and paper) and apply the model to an input such as ecg data. To the extent Applicant argues “Indeed, if there were no requirement for a reasonable degree of accuracy, the evaluation would not be based on the artificial ECG data but instead on the physician's evaluation of the single, existing ECG trace (since a physician would rely on that known data versus made-up, inaccurate data).”, this is a straw man argument. It is not arguing against the capability of the person to perform it, but is arguing that they wouldn’t do it that way because it is less practical or easier to do it another way. Just because one could do it another way doesn’t mean the person is incapable of going through the steps claimed. Applicants next argue the human mind is not capable not equipped to perform operations on voltage data. The discussion related to the “receiving” (or the “input” data) is part of the additional elements which are extra solution activity. Additionally, they are reading in elements which are not claimed such as elements of sampling frequency etc. The claims recite “receiving” and “input”, but Examiner notes that the data “received” could be from a hard drive on a remote/cloud server which could be from a decade ago or more (the input data can be from an old data set as well, Examiner notes it’s well known to divide datasets into subsets used for training, validation and test data). A user could review printouts of data to perform the recited functions, to provide meaningful insight. Applicant next argue that Examiners response reciting “there is nothing in the claim which would render . . .” ignores the fact the “generating . . .” step of the claim. Examiner disagrees. Again, Applicants are relying on elements not claimed. Contrary to Applicants argument there is nothing that stops the “generating one or more predictions . . .”, the predictions are based on the data input and the data input is just input data without limitation beyond being “second ECG data” (examiner notes that both the training data and the input data are “second ECG data”). Applicants are reading in the data includes data from a user currently. Phrased differently, while Applicants argued limitations may be encompassed under the BRI, the BRI is not limited to them. could be based on old data from an old dataset. Lastly Applicants restate a summary of their position based on arguments discussed above and Examiner find them not persuasive for the same reasons discussed above. Applicants next argue the amendments are sufficient to overcome the 101 (Examiner agrees there is support in [0079]). Examiner disagrees. The amendment in simpler terms recites taking the risk model (“first” AI model claimed) and first ECG data, using trial and error to determine what lead(s) from the first ECG data is/are the closest to the second ECG data by iteratively swapping in the second ECG data in for lead(s) of the first ECG data running each of the iterations through the first model with the swapped second ECG data and comparing the outputs of first model. As discussed above, Examiner believes a human mind is capable of applying the data to the model and determining the outcome, neither swapping the data which is input nor performing a comparison/“evaluating performance” of the iterations are interpreted as not being able to be performed in the mind. As such Examiner dose not find this persuasive. In conclusion, the arguments are not persuasive and the recited elements abstract ideas are mental process and the claims are directed to the abstract idea. Applicants next argue that it is integrated into a practical application by providing an improvement in the technical field. Applicants argue it is an improvement because it “translates” (i.e. what Applicants argue is the “new data”) the input to a larger set of leads, which seems to the abstract idea itself and thus not a practical application. Applicant points to “new data” however, this does not appear to be analogous to the “new data” referenced in Electric Power Group v. Alstom, S.A., 830 F.3d 1350, 1353-54, 119 USPQ2d 1739, 1741-42 (Fed. Cir. 2016). Applicants next argue that the addition of feeding the output of the “translating” model into the disease state model to get a disease state renders it an improvement, Examiner disagrees. These seem to be both stating the advantage is the abstract idea itself, which is not persuasive. Applicants argue that the Examiners arguments in the previous action “disregards two aspects of the practical application analysis . . .”, Examiner disagrees. First, Applicants more specifically seem to be arguing that the two models and the particular device are additional elements which aren’t considered. However, they are considered, while the processor, memory, “particular device” (wearable in WRC analysis) and ECG “including a predetermined number of leads” are interpreted as additional elements, the models are not being interpreted as additional elements but part of the abstract idea. All of which are accounted for in the rejection either as part of the extra solution activity of data gathering or the part of the rejection discussion “computing steps are recited at a high-level of generality . . .”. Thus, this is not persuasive. Second, they are arguing the elements aren’t considered as a whole, but the abstract idea(s) were considered in view of the additional elements and were not considered to be integrated into a practical application. The remaining arguments rely on the same argument discussed above and are not persuasive for the same reasons. 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 11-18, 20-28, 30 are rejected under 101 see the analysis below. Step 1 The claimed invention in claim 11 and 21 are directed to statutory subject matter as the claims recite a system and CRM. Step 2A, Prong 1 Regarding Claim 11 and 21, the recited step of “identify a first artificial intelligence model for a particular cardiac disease state; train the first artificial intelligence model on the first ECG data; . . . select, from the predetermined set of leads, a training lead or combination of leads that most closely resembles the at least one lead of the particular device, at least in part by iteratively replacing data from one or more leads of the predetermined set of leads with ECG data associated with the second number of leads and evaluating performance of each iteration of the first artificial intelligence model; train a second artificial intelligence model using a training dataset comprising ECG data including the predetermined set of leads, the second artificial intelligence model being configured to generate artificial EGG data corresponding to each lead of the predetermined set of leads based on the training lead or combination of leads; input the second ECG data into the trained second artificial intelligence model; translate, by the second artificial intelligence model, the second ECG data into artificial ECG data corresponding to each of the leads of the predetermined set of leads; and process, by the first artificial intelligence model, the artificial ECG data to generate one or more predictions regarding the particular cardiac disease state.” is directed to a mental process of performing concepts in the human mind (or by a human using the aid of pen and paper) and/or mathematical concept. That is nothing in the claim element precludes the step from practically being performed in the mind. For example, a human being can mentally review ECG data over time and associated disease states, training a neural network in the associations between different leads as well as disease states based on the leads allowing them to artificially generate (or inference) additional ECG leads and determine disease states based on the multiple lead ECG’s. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind, then it falls within the “Mental Processes” grouping of abstract ideas. Accordingly, the claim recites an abstract idea. Step 2A, Prong 2 Regarding Claims 11 and 21, the judicial exception is not integrated into a practical application. The claim includes the additional elements of “receive first electrocardiogram (ECG) data associated with a plurality of subjects and an electrocardiogram configuration including a predetermined set of leads and a time interval, the predetermined set of leads defining a first number and configuration of leads, the first electrocardiogram data comprising, for each lead included in the predetermined set of leads, voltage data associated with at least a portion of the time interval, . . . receive second ECG data derived from a particular device, the second ECG data being associated with a second number of leads, the second number of leads being fewer leads than the first number of leads;” amounts to insignificant, extra-solution activity in that the it is data gathering. The steps of “receiving . . .” amounts to insignificant, extra-solution activity data gathering. The processor (i.e., “processor”, “computer processor”, “cloud-computing device”, “server”, “watcher terminal”, “mobile device”, “user device”) in computing steps are recited at a high-level of generality (i.e., as a generic processor performing a generic computer function of determining outputs from inputs) such that it amounts no more than mere instructions to apply the exception using a generic computer component. Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea. The claim is directed to an abstract idea. Step 2B Regarding Claim 14, the claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As with step 2A, Prong 2 above, the claim includes the additional elements of “receive first electrocardiogram (ECG) data associated with a plurality of subjects and an electrocardiogram configuration including a predetermined set of leads and a time interval, the predetermined set of leads defining a first number and configuration of leads, the first electrocardiogram data comprising, for each lead included in the predetermined set of leads, voltage data associated with at least a portion of the time interval, . . . receive second ECG data derived from a particular device, the second ECG data being associated with a second number of leads, the second number of leads being fewer leads than the first number of leads;” amounts to insignificant, extra-solution activity in that the it is data gathering. The steps of “receiving . . .” amounts to insignificant, extra-solution activity data gathering. The processor (i.e., “processor”, “computer processor”, “cloud-computing device”, “server”, “watcher terminal”, “mobile device”, “user device”) in computing steps are recited at a high-level of generality (i.e., as a generic processor performing a generic computer function of determining outputs from inputs) such that it amounts no more than mere instructions to apply the exception using a generic computer component. Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea. The claim is directed to an abstract idea. Additionally, per the Berkheimer requirement, the wearable (Claims 4-7/14-17/24-27) devices, ECG sensing devices (with varying number of leads), processor/memory: (1) Alb see citations below shows the ECG sensing and processor/memory see the cited portions below, for example “Varying numbers of leads can be used to take the ECG, and different combinations of electrodes can be used to form the various leads.” [0049] and for the 12 lead version “The standard or conventional 12-lead ECG configuration uses 10 electrodes” [0051]. Examiner notes that if it is “standard or conventional” it is recognized as well known; (2) US 20140288435 to Richards et al. Fig. 1, [0078], [0081]-[0083], [0241]; (3) US 20150351690 see [0112]-[0116]; (4) Applicants background section recites “Large datasets with matched ECG and clinical health data come from healthcare providers and are predominantly 10-second 12-lead ECG devices. Portable clinical ECGs and portable consumer devices can collect ECG data more often but are often limited to 1 or a few leads.” [5] Which appears to be acknowledging that 10 second 12 lead ECG devices are known as are portable devices with fewer leads. As such the elements are shown to be WRC. The claim limitations when viewed individually and in combination therefore do not amount to significantly more than the abstract idea itself. The claims are therefore ineligible. Claims 12-18, 20, 22-28, 30 -32only further define the data gathering (insignificant, extra-solution activity) or the decisions made with the gathered data (i.e., only further define the mental process or mathematical concept). Therefore, the claims do not include any additional elements that show integration into a practical application and do not include any additional elements that amount to significantly more than the abstract idea. The claims are ineligible. Conclusion The prior art made of record and not relied upon is considered pertinent to applicant's disclosure: US 11576623 discusses generating ECG data US 20240341661 discusses generating ECG data LEE JAEHYEOK ET AL: "Reconstruction of Precordial Lead Electrocardiogram From Limb Leads Using the State-Space Model", IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS, IEEE, PISCATAWAY, NJ, USA, vol. 20, no. 3, 1 May 2016 (2016-05-01), pages 818-828, XP011609749, ISSN: 2168-2194, DOI: 10.1109/JBHI. 2015.2415519, viewed on 9/26/22; WANG LU-DI ET AL: "A novel method based on convolutional neural networks for deriving standard 12-lead ECG from serial 3-lead ECG", FRONTIERS OF INFORMATION TECHNOLOGY & ELECTRONIC ENGINEERING, ZHEJIANG UNIVERSITY PRESS, HEIDELBERG, vol. 20, no. 3, 19 April 2019 (2019-04-19), pages 405-413, XP036814011, ISSN: 2095-9184, DOI: 10.1631 FITEE.1700413, https://link.springer.com/article/10.1631/FITEE.1700413, viewed on 9/26/22; ZHANG QINGXUE ET AL: "All-ECG: A Least-number of Leads ECG Monitor for Standard 12-lead ECG Tracking during Motion*", 2019 IEEE HEALTHCARE INNOVATIONS AND POINT OF CARE TECHNOLOGIES, (HI-POCT), IEEE, 20 November 2019 (2019-11-20), pages 103-106, XP033691943, DOI: 10.1109/HI-POCT45284.2019.8962742, https://ieeexplore.ieee.org/document/8962742, viewed on 9/26/22; XU ZIJIAN ET AL: "Reconstruction of 12-Lead Electrocardiogram Based on GVM", 2018 SIXTH INTERNATIONAL CONFERENCE ON ADVANCED CLOUD AND BIG DATA (CBD), IEEE, 12 August 2018 (2018-08-12), pages 275-280, XP033443567, DOI: 10.1109/CBD.2018.00056, https://ieeexplore.ieee.org/document/8530852, viewed on 9/26/22; HUSSEIN ATOUI ET AL: "A Novel Neural-Network Model for Deriving Standard 12-Lead ECGs From Serial Three-Lead ECGs: Application to Self-Care", IEEE TRANSACTIONS ON INFORMATION TECHNOLOGY IN BIOMEDICINE, IEEE SERVICE CENTER, LOS ALAMITOS, CA, US, vol. 14, no. 3, 1 May 2010 (2010-05-01), pages 883-890, XP011345713, ISSN: 1089-7771, DOI: 10.1109/TITB.2010.2047754, https://ieeexplore.ieee.org/document/5445045, viewed on 9/26/22; Joon‐Myong Kwon et al. Deep Learning–Based Algorithm for Detecting Aortic Stenosis Using Electrocardiography. 21 Mar 2020. https://doi.org/10.1161/JAHA.119.014717. J Am Heart Assoc. 2020;9:e014717. DOI: 10.1161/JAHA.119.014717, viewed on 9/26/22 – recites the training and application of a 12 lead ECG to detect Aortic Stenosis; Tomer Golan et al. (2020). Improving ECG Classification Using Generative Adversarial Networks. Proceedings of the AAAI Conference on Artificial Intelligence, 34(08), 13280-13285. https://doi.org/10.1609/aaai.v34i08.7037. viewed on 9/27/22. I. Tomašić et al., "Electrocardiographic Systems With Reduced Numbers of Leads—Synthesis of the 12-Lead ECG," in IEEE Reviews in Biomedical Engineering, vol. 7, pp. 126-142, 2014, doi: 10.1109/RBME.2013.2264282. Jacob Gildenblat, Pruning deep neural networks to make them fast and small, 8/22/19, https://web.archive.org/web/20190822211756/https://jacobgil.github.io/deeplearning/pruning-deep-learning, viewed on 8/19/2024 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 JAMES R MOSS whose telephone number is (571)272-3506. The examiner can normally be reached Monday - Friday (9:30 am - 5:30 pm). 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, James Kish can be reached at (571) 272-5554. 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. /James Moss/Examiner, Art Unit 3792 /ALLEN PORTER/Primary Examiner, Art Unit 3796
Read full office action

Prosecution Timeline

Jul 21, 2022
Application Filed
Sep 27, 2022
Non-Final Rejection — §101
Feb 01, 2023
Response Filed
Jul 21, 2023
Final Rejection — §101
Oct 27, 2023
Request for Continued Examination
Nov 06, 2023
Response after Non-Final Action
Nov 28, 2023
Non-Final Rejection — §101
Feb 12, 2024
Interview Requested
Feb 27, 2024
Examiner Interview Summary
Feb 27, 2024
Applicant Interview (Telephonic)
May 06, 2024
Response Filed
Aug 20, 2024
Final Rejection — §101
Nov 21, 2024
Response after Non-Final Action
Dec 05, 2024
Response after Non-Final Action
Feb 24, 2025
Request for Continued Examination
Feb 26, 2025
Response after Non-Final Action
May 17, 2025
Non-Final Rejection — §101
Aug 21, 2025
Response Filed
Jan 07, 2026
Final Rejection — §101
Mar 02, 2026
Interview Requested
Mar 04, 2026
Applicant Interview (Telephonic)

Precedent Cases

Applications granted by this same examiner with similar technology

Patent 12589053
SYSTEM FOR DETECTING POSITION OF DISTAL END OF MEDICAL TUBE
2y 5m to grant Granted Mar 31, 2026
Patent 12558037
TECHNIQUES FOR USING DATA COLLECTED BY WEARABLE DEVICES TO CONTROL OTHER DEVICES
2y 5m to grant Granted Feb 24, 2026
Patent 12544587
PHOTOTHERAPEUTIC SYSTEMS INCLUDING SPREADING AND COLLIMATING FEATURES AND RELATED TECHNOLOGY
2y 5m to grant Granted Feb 10, 2026
Patent 12533258
SURGICAL TREATMENT FOR GLAUCOMA
2y 5m to grant Granted Jan 27, 2026
Patent 12527525
COMMUNICATION METHOD FOR COMMUNICATING MONITORING DATA AND MONITORING SYSTEM
2y 5m to grant Granted Jan 20, 2026
Study what changed to get past this examiner. Based on 5 most recent grants.

AI Strategy Recommendation

Get an AI-powered prosecution strategy using examiner precedents, rejection analysis, and claim mapping.
Powered by AI — typically takes 5-10 seconds

Prosecution Projections

7-8
Expected OA Rounds
51%
Grant Probability
92%
With Interview (+41.0%)
3y 3m
Median Time to Grant
High
PTA Risk
Based on 261 resolved cases by this examiner. Grant probability derived from career allow rate.

Sign in with your work email

Enter your email to receive a magic link. No password needed.

Personal email addresses (Gmail, Yahoo, etc.) are not accepted.

Free tier: 3 strategy analyses per month