DETAILED ACTION
Status of Claims
This action is in reply to the communication filed on 7 April, 2026.
Claims 1, 6, 15 and 20 have been amended.
Claims 1 – 20 are currently pending and have been examined.
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 .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 7 April, 2026 has been entered.
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.
Claims 1 – 20 are rejected under 35 U.S.C. 112(b) as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor regards as the invention. Claims 1 and 15 recite a sequence of steps related to training machine learning models, including:
inputting the prediction target and the clinical data into machine learning models that are trained to predict diagnosis risk, wherein the machine learning models include (i) a first model trained from existing data and (ii) a second model that is distinct from the first model, and the first model is configured to determine the diagnosis risk score independent of the training of the second model;
training the second model of the machine learning models using the screening result;
wherein the machine learning models are trained using training data comprising historical claims data, historical clinical data, and historical demographic data, from a population of prior patients, and
wherein the machine learning models are trained to detect correlation between medical diagnosis signals identified from the training data, and a positive result from the screening mechanism for likelihood of the medical diagnosis.
Examiner cannot determine the metes and bounds of the claims. In particular, it is unclear what training data is used to train the models and what inputs the models require. For example, the claims require training a first and second model. The first model is trained using “existing data” – i.e. training data comprising historical claims data, historical clinical data, and historical demographic data, from a population of prior patients.” The first model is trained to detect “correlation between medical diagnosis signals identified from the training data, and a positive result from the screening mechanism.” The first model only receives as input the “prediction target and the clinical data”.
Similarly, the claims recite that the second model is trained using training data comprising historical claims data, historical clinical data, and historical demographic data, from a population of prior patients.” The second model is also trained to detect “correlation between medical diagnosis signals identified from the training data, and a positive result from the screening mechanism.” The second model is further trained using the screening result.
The specification discloses that the first model (i.e. a base or initial model) is trained to determine a diagnosis risk score using historical claims, clinical and demographic training data, and the prediction target alone; and the second model is trained using historical claims, clinical and demographic data as well as the screening result. In particular, the recited second model is disclosed as the first model being enhanced using reinforcement learning to refine the first model using the screening results.
Further, the specification discloses that the screening results are obtained using well-known questionnaires; and that the questionnaires are provided to patient based on their diagnosis risk score. As such, it appears that the “transmitting . . . a recommendation” step should occur before the “performing a screening” step. The recommendation for further evaluation IS a recommendation to perform the screening mechanism.
Examiner suggest the following sequence of steps:
A method of predicting a medical diagnosis for a patient, independent of prior diagnosis obtained from interviewing or examining the patient, the method comprising:
autonomously receiving, at a processor of a computer system, claims data, clinical data and demographic data relating to the patient, from one or more network databases;
determining, by the processor, from the claims data, whether a prediction target for the medical diagnosis is present;
in response to a determination that the prediction target is present:
inputting the prediction target and the clinical data into machine learning models that are trained to predict diagnosis risk, wherein the machine learning models include (i) a first model trained from existing data and (ii) a second model that is distinct from the first model, and the first model is configured to determine the diagnosis risk score independent of the training of the second model;
determining, using the first machine learning model, a diagnosis risk score;
wherein the first machine learning model is trained using training data comprising historical claims data, historical clinical data, and historical demographic data, from a population of prior patients, to detect correlation between medical diagnosis signals identified from the training data, and a positive result from a screening mechanism for likelihood of the medical diagnosis, and
determining, by the processor, a care seeking propensity score, from the demographic data, wherein the care seeking propensity score is related to whether the patient is a member of a group with a propensity to seek care that is lower than a reference care seeking propensity score for other patients;
weighting, by the processor, the diagnosis risk score by the care seeking propensity score to create a weighted diagnosis risk score;
determining whether the weighted diagnosis risk score indicates a likelihood of the medical diagnosis; and
in response to the determination that the weighted diagnosis risk score indicates a likelihood of the medical diagnosis:
automatically transmitting, over a network, a recommendation for further patient screening to a digital device associated with the patient;
performing a screening mechanism on the patient to obtain a screening result;
training the second model of the machine learning models using the screening result;
wherein training the second machine learning model comprises performing reinforcement learning applied to the first machine learning model based on the screening results.
Response to Arguments
Applicant’s arguments, filed 7 April, 2026, have been fully considered and are persuasive. The rejections have been withdrawn.
The U.S.C. §101 Rejection
Applicant argues that the architecture of the machine learning models provides an improvement, although there is no support for some of the specific improvements asserted by the Applicant (i.e. “balance the demand for computational resources” “reduce processing time”). Nonetheless, the claims (as suggested above) are directed to a process that provides an improvement to a machine learning model itself, by applying reinforcement learning to a trained model using actual results (i.e. ground truth). Even though reinforcement learning may be routine and purely conventional, the specification expressly discloses an improved model using this technique.
CONCLUSION
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
US 11,482,322 B1 to Bhansali et al. discloses a system and method for medical risk determination that includes using machine learning models and feedback for refining the model.
Any inquiry of a general nature or relating to the status of this application or concerning this communication or earlier communications from the Examiner should be directed to John A. Pauls whose telephone number is (571) 270-5557. The Examiner can normally be reached on Mon. - Fri. 8:00 - 5:00 Eastern. If attempts to reach the examiner by telephone are unsuccessful, the Examiner’s supervisor, Robert Morgan can be reached at (571) 272-6773.
Official replies to this Office action may now be submitted electronically by registered users of the EFS-Web system. Information on EFS-Web tools is available on the Internet at: http://www.uspto.gov/patents/process/file/efs/guidance/index.jsp. An EFS-Web Quick-Start Guide is available at: http://www.uspto.gov/ebc/portal/efs/quick-start.pdf.
Alternatively, official replies to this Office action may still be submitted by any one of fax, mail, or hand delivery. Faxed replies should be directed to the central fax at (571) 273-8300. Mailed replies should be addressed to “Commissioner for Patents, PO Box 1450, Alexandria, VA 22313-1450.” Hand delivered replies should be delivered to the “Customer Service Window, Randolph Building, 401 Dulany Street, Alexandria, VA 22314.”
/JOHN A PAULS/Primary Examiner, Art Unit 3683
Date: 29 June, 2026
Prosecution Insights