| CPC G16H 20/13 (2018.01) [G16H 10/40 (2018.01); G16H 10/60 (2018.01); G16H 20/10 (2018.01); G16H 10/20 (2018.01); G16H 15/00 (2018.01); G16H 70/20 (2018.01)] | 20 Claims |
|
1. A system for modifying operation of a medical system of a computer network to enhance patient safety and system efficiency, comprising:
a processor for executing an application programming interfaces (API) communicatively coupled to a service layer of an electronic health record (EHR) system of the medical system, wherein the API is configured to execute a plurality of operational functions of the medical system comprising patient data access and prescription processing;
a memory, operatively coupled to the processor;
a communications interface, operatively coupled to the processor, wherein the communications interface is configured to
(i) communicate with an electronic prescription application to receive prescription drug monitoring program (PDMP) data from the computer network,
(ii) communicate with a database to receive medical data and prescription data associated with one or more patients, and
(iii) receive image data of the one or more patients from an image capturing device; and
a learning logic, configured to
(i) electronically extract contextual knowledge data comprising summarization information from the received medical data, and extract structural knowledge data comprising instances and/or labels ontology from the received prescription data,
(ii) apply a first computer learning model to the contextual knowledge data and structural knowledge data to generate one or more predictive model values, based on a structured combination of the contextual knowledge data and structural knowledge data, the predictive model values indicating a potential drug abuse condition,
(iii) electronically process the image data to determine a feature map, and extract features from the feature map if the one or more predictive model values meet or exceed a configured threshold,
(iv) electronically process the extracted features to determine one or more classifications for the image data, the classifications including one or more facial emotive expressions indicative of the potential drug abuse condition, and
(v) apply a second computer learning model using the one or more classifications and predictive model values to generate a second predictive model value confirming the potential drug abuse condition,
wherein the processor is configured to generate and transmit executable code based on the second predictive model value via the communications interface to the computer network, the executable code being configured to modify the operation of the medical system by enabling or disabling one or more of the configured operational functions of the API, comprising selectively restricting patient data access or altering prescription processing workflows to mitigate the potential drug abuse condition, thereby improving patient safety and reducing network resource consumption.
|