Computer network architecture with benchmark automation, machine learning and artificial intelligence for measurement factors
The present disclosure is related generally to computer network architectures for machine learning, and more specifically, to computer network architectures for the automated production and distribution of custom healthcare performance benchmarks for specific patient cohorts. Embodiments allow specification and automated production of benchmarks using any of many dozens of patient, disease process, facility, and physical location attributes. Embodiments may use an analytic module web application and a benchmark service module web application, with other architecture components. Embodiments may include a combination of third-party databases to generate benchmarks and to drive the forecasting models, including social media data, financial data, socio-economic data, medical data, search engine data, e-commerce site data, and other databases.
1. A computer network architecture with benchmark automation, artificial intelligence and machine learning, comprising:
a prediction module with a prediction generator and updated system databases
a learning module with a training submodule, in electronic communication with the prediction module and the updated system database,
a prediction web application in electronic communication with the prediction module and a user device, and
an analytic module web application (AMWA) in electronic communication with the prediction module, the prediction web application, and a user device, and
a benchmark service module web application (BSMWA) in electronic communication with the AMWA and the user device, and
wherein, the AMWA is configured to:
a. receive a user request for analysis and performance benchmarks, with patient clinical episode data,
b. analyze the patient clinical episode data,
c. identify underperforming patient cohorts, and
d. request the BSMWA to generate performance benchmarks for the identified cohorts, and
wherein the BSMWA is configured to:
a. identify updated system databases and third-party databases with relevant patient claims data,
b. generate and distribute requests for data to the identified databases,
c. receive the requested data and store the data received from the user requests and the data requests in the updated system databases,
d. generate a benchmark response report for the identified patient cohorts using the received data,
e. transfer the benchmark response report to the user device, and
f. access the prediction web application to predict the performance of the identified patient cohorts and impacts of meeting the benchmarks, and
wherein, the learning module is configured to:
receive a list of algorithm definitions and datasets for prediction,
automatically calibrate one or more defined algorithms with the identified updated system databases,
test the calibrated algorithms with a plurality of evaluation metrics,
store the tested algorithms and the evaluation metrics in a library,
automatically select a tested algorithm,
wherein the selected tested algorithm is utilized by the prediction web application to predict the performance of the identified patient cohorts and the impacts of meeting the benchmarks;
update further the identified updated system databases with third party data, and with user episode data, and
upon occurrence of an event or periodically, re-execute the calibrate, test, store, and select steps after the update of the identified updated system databases step.
2. The computer network architecture in claim 1 , wherein the identified databases include data that is from a party that is a member of the group consisting of:
hospitals,
medical practices,
insurance companies,
credit reporting agencies, and
credit rating agencies, and
the identified databases include patient medical data, patient personal data, patient outcome data, and medical treatment data;
the user is a member of the group comprising:
hospitals,
medical practices, and
insurance companies.
3. The computer network architecture of claim 1 wherein the user device is remote from the AMWA and the BSMWA, and
the user device is a member of the group consisting of:
a computer,
a desktop PC,
a laptop PC,
a smart phone,
a tablet computer, and
a personal wearable computing device.
4. The computer network architecture of claim 1 , wherein the BSMWA communicates with the third-party databases by the Internet, or an extranet, or a VPN, or other network, and the AMWA and BSMWA are generic for any user, or customized for a specific user, or class of user.
5. The computer architecture in claim 1 , wherein the identified databases contain data from at least one third party, containing data of a plurality of types consisting of: medical claims data, prescription refill data, publicly available social media data, credit agency data, marketing data, travel website data, e-commerce website data, search engine data, credit card data, credit score and credit history data, lending data, mortgage data, financial data, travel data, geolocation data, and telecommunications usage data.
6. A computer network architecture with benchmark automation, artificial intelligence and machine learning, comprising:
a prediction module with a prediction generator and updated system databases,
a learning module with a training submodule, in electronic communication with the prediction module and the updated system database,
a prediction web application in electronic communication with the prediction module and a user device, and
an analytic module web application (AMWA) in electronic communication with the prediction module, the prediction web application, and a user device, and
a benchmark service module web application (BSMWA) in electronic communication with the prediction module, the prediction web application, the AMWA and the user device, and
wherein, the AMWA is configured to:
a. receive a user request for analysis and performance benchmarks,
b. analyze patient clinical episode data,
c. identify underperforming patient cohorts, and
d. request the BSMWA to generate performance benchmarks for the identified cohorts, and
wherein the BSMWA is configured to:
a. identify updated system databases and third-party databases with relevant patient claims data,
b. generate and distribute requests for data to the identified databases,
c. receive the requested data and store the received data in the updated system databases,
d. generate a benchmark response report for the identified patient cohorts using the received data, and
e. transfer the benchmark response report to the user device; and
f. access the prediction web application to predict the performance of the identified patient cohorts and impacts of meeting the benchmarks, and
wherein the identified databases include data that is from a party that is a member of the group consisting of:
hospitals,
medical practices,
insurance companies,
credit reporting agencies, and
credit rating agencies, and
the identified databases include patient medical data, patient personal data, patient outcome data, and medical treatment data; and
the user is a member of the group comprising:
hospitals,
medical practices, and
insurance companies; and
wherein the user device is remote from the AMWA and the BSMWA, and
the user device is a member of the group consisting of:
a computer,
a desktop PC,
a laptop PC,
a smart phone,
a tablet computer, and
a personal wearable computing device; and
wherein the BSMWA communicates with the third-party databases by the Internet, or an extranet, or a VPN, or other network, and the AMWA and BSMWA are generic for any user, or customized for a specific user, or class of user; and
wherein the identified databases contain data from at least one third party, containing data of a plurality of types consisting of: medical claims data, prescription refill data, publicly available social media data, credit agency data, marketing data, travel website data, e-commerce website data, search engine data, credit card data, credit score and credit history data, lending data, mortgage data, financial data, travel data, geolocation data, and telecommunications usage data; and
wherein, the learning module is configured to:
receive a list of algorithm definitions and datasets for prediction,
automatically calibrate one or more defined algorithms with the identified updated system databases,
test the calibrated algorithms with a plurality of evaluation metrics,
store the tested algorithms and the evaluation metrics in a library,
automatically select a tested algorithm,
wherein the selected tested algorithm is utilized by the prediction web application to predict the performance of the identified patient cohorts and the impacts of meeting the benchmarks;
update further the identified updated system databases with third party data, and with user episode data, and
upon occurrence of an event or periodically, re-execute the calibrate, test, store, and select steps after the update of the identified updated system databases step.