CTNF 18/746,289 CTNF 87828 07-03-aia AIA 15-10-aia The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA. DETAILED ACTION This communication is responsive to the application filed on 06/18/2024. Claims 1-8 are pending in this application. This action is made non-final . Claim Rejections - 35 USC § 101 07-04-01 AIA 07-04 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 1-8 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. Step 1: Statutory Category The claim is directed to a “learning system” comprising a processor and memory. Accordingly, it falls within a statutory category (machine). Step 2A, Prong 1: Whether the Claim Recites a Judicial Exception The claim recites steps of: receiving patient-related data, training multiple machine learning models, computing performance scores, and selecting a model based on those scores for predicting hospital length of stay. This is about collecting data, analyzing it using mathematical models, and making a selection. data collection + analysis + result selection = abstract idea The claim recites an abstract idea. Step 2A, Prong 2: Does the claim integrates the abstract idea into a practical application? The “learning system” uses generic processor and memory. There is no specific: improvement to computer functionality, or technical solution to a technical problem. The additional limitation “in consideration of available computational resources and time budget” lacks implementation details. There is no integration into a practical application such as improving hospital system operation at a systems level. The claim does not integrate the abstract idea into a practical application. Step 2B: Inventive Concept (Significantly More) The remaining elements: processor memory generic ML training They are well-understood and routine. There is no element that provides a technical improvement. “apply machine learning to patient data and pick the best model” is a generic instruction to apply an abstract idea using conventional technology. Hence, the claim does not include an inventive concept. Claims 2-6 are rejected under 35 USC § 101 because the additional limitations do not integrate the abstract idea into a practical application and do not amount to significantly more than the abstract idea itself. The additional elements merely recite well-understood, routine, and conventional computer components performing their ordinary function. Claim Rejections - 35 USC § 103 07-06 AIA 15-10-15 In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status . 07-20-aia AIA 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. 07-21-aia AIA Claim s 1 and 7-8 are rejected under 35 U.S.C. 103 as being unpatentable over Thomas et al. (US 2021/0098090; Hereinafter Thomas) in view of J et al (US 2020/0159690; hereinafter J) . Re claims 1 and 7-8, Thomas teaches a learning system comprising: at least one memory configured to store instructions (fig. 17, system memory); and at least one processor configured to execute the instructions (fig. 17, CPU) to: receive at least health data of patients ([0046], patient medical records (e.g., EHRs), patient admission information, patient insurance information, patient demographic information, patient family support information, case worker impression information, clinician reports/notes, imaging study information, laboratory information, tracked patient status information, tracked patient vital signs information, tracked clinical event information (e.g., procedures performed, medications administered, etc.), tracked patient location information, and the like. Also see fig. 2 and [0062], the patient medical history information); train a plurality of machine learning models based on the at least one of the health data of patients ([0033], predicting patient care outcomes employ one or more machine learning models respectively trained to predict the patient care outcome based on learned correlations between a variety of unique combinations of clinical and non-clinical factors. For example, the clinical factors can include information regarding the patient's medical history prior to admission); Thomas teaches training machine learning model ([0033]) but Thomas does not explicitly teach in consideration of available computational resources and time budget but it is taught by J ([0003], to train machine-learning systems may be costly, but necessary given various constraints (e.g., limited time, resources, and so on) that are typically applied to this kind of work). Thomas does not teach compute performance scores of the plurality of trained machine learning models but it taught by J ([0045], Lead intelligence uses the machine learning model trained on past data to predict the probability of a lead getting converted into an opportunity. Lead intelligence is a ranking technique. Higher the score means higher the chance of getting the lead converted into an opportunity. Lead intelligence prioritizes leads based on the propensity to be converted into an opportunity. It helps sales and marketing teams predict and prioritize leads that are likely to get converted into successful opportunities compared to opportunities that are unlikely to get converted); and Thomas teaches select a machine learning model used for a prediction of a length of stay in a hospital for a patient among the plurality of trained machine learning models ([0004], employs one or more forecasting machine learning models to predict care outcomes for the complex patients based on the patient information, including remaining LOS in the hospital, discharge destinations and their likelihoods, readmission risks and safety risks. The computer executable components further comprise a reporting component that provides patient care outcomes information to one or more care providers to facilitate managing and coordinating inpatient and post-discharge care for the respective patients, wherein patient care outcomes information identifies the complex patients, their length of stay, their discharge destinations) but Thomas does not teach based on the computed performance scores. However, it is taught by J ([0045], Lead intelligence uses the machine learning model trained on past data to predict the probability of a lead getting converted into an opportunity. Lead intelligence is a ranking technique. Higher the score means higher the chance of getting the lead converted into an opportunity. Lead intelligence prioritizes leads based on the propensity to be converted into an opportunity. It helps sales and marketing teams predict and prioritize leads that are likely to get converted into successful opportunities compared to opportunities that are unlikely to get converted. Also see [0074], machine-learning scenario content, such as opportunity score, lead scoring, and so on, is pre-delivered to customers for activation. Upon activation, data is automatically extracted from target tables using an auto-awareness technique. A feature engineering framework generates the necessary features automatically. These features are then passed to a back-end service (e.g., SAP PAi) for model selection and deployment. Using techniques of auto-sampling, the best performing model is picked and a probabilistic score is applied to all new records in the system (e.g., using a batch model)) . Therefore, it would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to add the teaching as seen in J’s content into Thomas’s invention because it would improve predictive accuracy while efficiently utilizing computational resources and time as both references address machine learning model optimization . 07-21-aia AIA Claim 6 is rejected under 35 U.S.C. 103 as being unpatentable over Thomas in view of J and further in view of Zhang et al (US 2020/0175362; hereinafter Zhang) . Re claim 6, Thomas and J do not teach wherein a hyperparameter of the selected machine learning model is optimized by using Hierarchical Automated Machine Learning with Time-awareness (HAMLET). However, it is taught by Zhang ([0074], The system 500 uses AutoML if, for example, the accuracy on the new task does not meet desired accuracy requirements/standards, for example, the model size/capacity may be too small to accommodate two tasks. The system 500 may then expand the deep learning model size, for example, according to the algorithm in Table 2. Third, when the accuracy requirement/standards are met and the model size has been increased significantly, the system 500 can compress the adapted model (e.g., using the compression according to equation 11 above and the algorithm in Table 1 and step 7). As a result, the computation, memory, and power cost on the adapted model is not greatly increased or is maintained). Therefore, it would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to add the teaching as seen in Zhang’s content into J and Thomas’s invention because it would improve predictive accuracy and reduce computational and training time, yielding predictable results. Allowable Subject Matter Claims 2-5 are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims because the additional limitations integrate the abstract idea into a practical application and therefore overcome the rejection under 35 USC § 101 . Conclusion The prior art made of record on form PTO-892 and not relied upon is considered pertinent to applicant's disclosure. Applicant is required under 37 C.F.R. § 1.111 ( c ) to consider these references fully when responding to this action. Any inquiry concerning this communication or earlier communications from the examiner should be directed to TOAN H VU whose telephone number is (571)270-3482. The examiner can normally be reached on PHP 9-5:30 PM. If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Stephen Hong can be reached on 571-272-4124. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300. Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000. /TOAN H VU/Primary Examiner, Art Unit 2178 Application/Control Number: 18/746,289 Page 2 Art Unit: 2178 Application/Control Number: 18/746,289 Page 3 Art Unit: 2178 Application/Control Number: 18/746,289 Page 4 Art Unit: 2178 Application/Control Number: 18/746,289 Page 5 Art Unit: 2178 Application/Control Number: 18/746,289 Page 6 Art Unit: 2178 Application/Control Number: 18/746,289 Page 7 Art Unit: 2178 Application/Control Number: 18/746,289 Page 8 Art Unit: 2178 Application/Control Number: 18/746,289 Page 9 Art Unit: 2178 Application/Control Number: 18/746,289 Page 10 Art Unit: 2178
Prosecution Insights