Computer network architecture with machine learning and artificial intelligence and patient risk scoring
Embodiments relate generally to computer network architectures for machine learning, and more specifically, to computer network architectures in the context of program rules, using combinations of defined patient clinical episode metrics and other clinical metrics, thus enabling superior performance of computer hardware. Aspects of embodiments herein are specific to patient clinical episode definitions, and are applied to the specific outcomes of highest concern to each episode type. Furthermore, aspects of embodiments herein produce more accurate and reliable predictions of possible patient outcomes and metrics.
1. A computer network architecture with artificial intelligence and machine learning, comprising:
one or more memory units configured to store one or more instructions,
one or more processors, and
a web application in electronic communication with both the one or more processors, and a user device,
wherein the web application is adapted to:
a. automatically receive a user prediction request for patient risk scoring, for a patient, for a specific clinical episode type,
b. automatically transmit the user prediction request for patient risk scoring to the one or more processors,
c. automatically calibrate a correlation of demographic, social and medical attributes of the patient, to an outcome of the specific patient clinical episode type to be analyzed,
d. automatically combine data of patient interactions with a healthcare team and proprietary third-party patient attributes data into model attributes,
e. automatically produce, for display to a treating clinician, an interpretable schedule, adjusted for program rules, of an impact of patient attributes on outcomes of a patient episode, the interpretable schedule including a prediction of total cost, after adjustment for the program rules, of episode treatment, wherein the interpretable schedule is a schedule of probabilities of each parameter of each possible patient episode outcome of interest for the request for patient risk scoring, the interpretable schedule indicating statistics of patient population demographic groups used to produce the interpretable schedule, the interpretable schedule providing no confidential data associated with individuals in the patient population demographic groups,
wherein the one or more processors are configured, upon executing the one or more instructions, to:
automatically receive a list of algorithm definitions and datasets for patient risk scoring,
automatically calibrate one or more defined algorithms with a database,
automatically test the calibrated algorithms with a plurality of evaluation metrics,
automatically store the calibrated algorithms and evaluation metrics in a library,
automatically select an algorithm for patient risk scoring based on evaluation metrics,
automatically update further the database with third party data, and with user episode data, and
automatically re-execute the calibrate, test, store, and select steps after the update of the database step,
wherein the one or more processors are further configured, upon executing the one or more instructions, to:
receive a user prediction request for patient risk scoring, including episode data and a client model,
run a currently selected algorithm corresponding to a user of the episode data, and
generate patient risk scoring prediction output,
generate a patient risk scoring prediction report based on an algorithm output, and
transmit the patient risk scoring prediction report to the user,
wherein the algorithm definitions are of types that are members of the group consisting of:
multi-level models,
random forest regression,
logistical regression,
gamma-distributed regression, and
linear regression;
the third party data are from a party that is a member of the group consisting of:
hospitals,
medical practices,
insurance companies,
credit reporting agencies, and
credit rating agencies;
the database includes patient medical data, patient personal data, patient outcome data, and medical treatment data;
the episode data include individual patient medical data and personal data; and the user is a member of the group consisting of:
hospitals,
medical practices, and
insurance companies,
wherein the user device is remote from the one or more processors, 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,
wherein the web application communicates with the user device by the Internet, or an extranet, or a VPN, or other network, and the web application is generic for any user, or customized for a specific user, or class of user, and
wherein the user prediction request requests calibration of the correlation of demographic, social and medical attributes of the patient, to the outcome of a specific patient clinical episode type.
2. A computer network architecture with artificial intelligence and machine learning, comprising:
one or more memory units configured to store one or more instructions,
one or more processors, and
a web application in electronic communication with both the one or more processors and a user device,
wherein the web application is adapted to:
a. automatically receive a user prediction request for patient risk scoring, for a patient, for a specific clinical episode type,
b. automatically transmit the user prediction request for patient risk scoring to the one or more processors,
c. automatically calibrate a correlation of demographic, social and medical attributes of a patient, to an outcome of the specific patient clinical episode type to be analyzed,
d. automatically combine data of patient interactions with a healthcare team and proprietary third-party patient attributes data into model attributes,
e. automatically produce, for display to a treating clinician, an interpretable schedule, adjusted for program rules, of an impact of patient attributes on outcomes of a patient episode, the interpretable schedule including a prediction of total cost, after adjustment for the program rules, of episode treatment, wherein the interpretable schedule is a schedule of probabilities of each parameter of each possible patient episode outcome of interest for the request for patient risk scoring, the interpretable schedule indicating statistics of patient population demographic groups used to produce the interpretable schedule, the interpretable schedule providing no confidential data associated with individuals in the patient population demographic groups.
3. The computer network architecture of claim 2 ,
wherein the one or more processors are configured, upon executing the one or more instructions, to:
automatically receive a list of algorithm definitions and datasets for patient risk scoring,
automatically calibrate one or more defined algorithms with a database,
automatically test the calibrated algorithms with a plurality of evaluation metrics,
automatically store the calibrated algorithms and evaluation metrics in a library,
automatically select an algorithm for patient risk scoring based on evaluation metrics,
automatically update further the database with third party data, and with user episode data, and
automatically re-execute the calibrate, test, store, and select steps after the update of the database step.
4. The computer network architecture of claim 3 ,
wherein the algorithm definitions are of types that are members of the group consisting of:
multi-level models,
random forest regression,
logistical regression,
gamma-distributed regression, and
linear regression;
the third party data are from a party that is a member of the group consisting of:
hospitals,
medical practices,
insurance companies,
credit reporting agencies, and
credit rating agencies;
the database includes patient medical data, patient personal data, patient outcome data, and medical treatment data.
5. The computer network architecture of claim 2 ,
wherein the one or more processors are configured, upon executing the one or more instructions, to:
receive a user prediction request for patient risk scoring, including episode data and a client model,
run a currently selected algorithm corresponding to a user of the episode data, and
generate patient risk scoring prediction output,
generate a patient risk scoring prediction report based on an algorithm output, and
transmit the patient risk scoring prediction report to the user.
6. The computer network architecture of claim 2 ,
wherein the user device is remote from the one or more processors, 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.
7. The computer network architecture of claim 2 , wherein the web application communicates with the user device by the Internet, or an extranet, or a VPN, or other network, and the web application is generic for any user, or customized for a specific user, or class of user, and
wherein the user prediction request requests calibration of the correlation of demographic, social and medical attributes of the patient, to the outcome of a specific patient clinical episode type.
8. A computer network architecture with a web application for artificial intelligence and machine learning, comprising:
one or more memory units configured to store one or more instructions,
one or more processors, and
a web application in electronic communication with both the one or more processors, and a user device, and
wherein the web application is adapted to:
a. automatically receive a user prediction request for patient risk scoring, for a patient, for a specific clinical episode type,
b. automatically transmit the user prediction request for patient risk scoring to the one or more processors,
c. automatically calibrate a correlation of demographic, social and medical attributes of the patient, to an outcome of the specific patient clinical episode type to be analyzed,
d. automatically combine data of patient interactions with a healthcare team and proprietary third-party patient attributes data into model attributes,
e. automatically produce, for display to a treating clinician, an interpretable schedule, adjusted for program rules, of an impact of patient attributes on outcomes of a patient episode, the interpretable schedule including a prediction of total cost, after adjustment for the program rules, of episode treatment, wherein the interpretable schedule is a schedule of probabilities of each parameter of each possible patient episode outcome of interest for the request for patient risk scoring, the interpretable schedule indicating statistics of patient population demographic groups used to produce the interpretable schedule, the interpretable schedule providing no confidential data associated with individuals in the patient population demographic groups,
wherein the one or more processors are configured, upon executing the one or more instructions, to:
automatically receive a list of algorithm definitions and datasets for patient risk scoring,
automatically calibrate one or more defined algorithms with a database,
automatically test the calibrated algorithms with a plurality of evaluation metrics,
automatically store the calibrated algorithms and evaluation metrics in a library,
automatically select an algorithm for patient risk scoring based on evaluation metrics,
automatically update further the database with third party data, and with user episode data, and
automatically re-execute the calibrate, test, store, and select steps after the update of the database step,
wherein the one or more processors are further configured, upon executing the one or more instructions, to:
receive a user prediction request for patient risk scoring, including episode data and a client model,
run a currently selected algorithm corresponding to a user of the episode data, and
generate patient risk scoring prediction output,
generate a patient risk scoring prediction report based on the algorithm output, and
transmit the patient risk scoring prediction report to the user,
wherein the algorithm definitions are of types that are members of the group consisting of:
multi-level models,
random forest regression,
logistical regression,
gamma-distributed regression, and
linear regression;
the third party data are from a party that is a member of the group consisting of:
hospitals,
medical practices,
insurance companies,
credit reporting agencies, and
credit rating agencies;
the database includes patient medical data, patient personal data, patient outcome data, and medical treatment data;
the episode data include individual patient medical data and personal data; and the user is a member of the group consisting of:
hospitals,
medical practices, and
insurance companies,
wherein the user device is remote from the one or more processors, 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,
wherein the web application communicates with the user device by the Internet, or an extranet, or a VPN, or other network, and the web application is generic for any user, or customized for a specific user, or class of user, and
wherein the user prediction request requests calibration of the correlation of demographic, social and medical attributes of the patient, to the outcome of a specific patient clinical episode type.
9. A computer network architecture with a web application with artificial intelligence and machine learning, comprising:
one or more memory units configured to store one or more instructions,
one or more processors, and
a web application in electronic communication with both the one or more processors, and a user device,
wherein the web application is adapted to:
a. automatically receive a user prediction request for patient risk scoring, for a patient, for a specific clinical episode type,
b. automatically transmit the user prediction request for patient risk scoring to the one or more processors,
c. automatically calibrate a correlation of demographic, social and medical attributes of the patient, to an outcome of the specific patient clinical episode type to be analyzed,
d. automatically combine data of patient interactions with a healthcare team and proprietary third-party patient attributes data into model attributes,
e. automatically produce, for display to a treating clinician, an interpretable schedule, adjusted for program rules, of an impact of patient attributes on outcomes of a patient episode, the interpretable schedule including a prediction of total cost, after adjustment for the program rules, of episode treatment, wherein the interpretable schedule is a schedule of probabilities of each parameter of each possible patient episode outcome of interest for the request for patient risk scoring, the interpretable schedule indicating statistics of patient population demographic groups used to produce the interpretable schedule, the interpretable schedule providing no confidential data associated with individuals in the patient population demographic groups.
10. The computer network architecture of claim 9 :
wherein the one or more processors are configured, upon executing the one or more instructions, to:
automatically receive a list of algorithm definitions and datasets for patient risk scoring,
automatically calibrate one or more defined algorithms with a database,
automatically test the calibrated algorithms with a plurality of evaluation metrics,
automatically store the calibrated algorithms and evaluation metrics in a library,
automatically select an algorithm for patient risk scoring based on evaluation metrics,
automatically update further the database with third party data, and with user episode data, and
automatically re-execute the calibrate, test, store, and select steps after the update of the database step.
11. The computer network architecture of claim 10 ,
wherein the algorithm definitions are of types that are members of the group consisting of:
multi-level models,
random forest regression,
logistical regression,
gamma-distributed regression, and
linear regression;
the third party data are from a party that is a member of the group consisting of:
hospitals,
medical practices,
insurance companies,
credit reporting agencies, and
credit rating agencies;
the database includes patient medical data, patient personal data, patient outcome data, and medical treatment data.
12. The computer network architecture of claim 9 ,
wherein the one or more processors are configured, upon executing the one or more instructions, to:
receive a user prediction request for patient risk scoring, including episode data and a client model,
run a currently selected algorithm corresponding to a user of the episode data, and
generate patient risk scoring prediction output,
generate a patient risk scoring prediction report based on the algorithm output, and
transmit the patient risk scoring prediction report to the user.
13. The computer network architecture of claim 9 ,
wherein the user device is remote from the one or more processors, 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.
14. The computer network architecture of claim 9 , wherein the web application communicates with the user device by the Internet, or an extranet, or a VPN, or other network, and the web application is generic for any user, or customized for a specific user, or class of user, and
wherein the user prediction request requests calibration of the correlation of demographic, social and medical attributes of the patient, to the outcome of a specific patient clinical episode type.