| CPC G16H 10/60 (2018.01) [G06F 17/11 (2013.01); G06N 20/00 (2019.01); H04L 67/01 (2022.05)] | 20 Claims |
|
1. A computer system, comprising:
a processing system comprising a processing device and memory;
computer storage medium accessible by the processing system and storing data structures representing health care information for a plurality of patients, the health care information including, for the plurality of patients, normalized data from a plurality of sources;
a plurality of classifiers, each classifier representing a respective different category associated with a respective medical condition, from among a plurality of such categories, each classifier comprising computer program instructions that, when processed by the processing system, configure the processing system to be comprising:
i. an input connected to receive health care information for a patient from the computer storage, and
ii. an output indicative of the likelihood the health care information for the patient indicates the patient has the medical condition associated with the category represented by the classifier,
wherein each classifier is based on training a model using a respective machine learning algorithm using a respective training set including at least some of the normalized data from the stored health care information for patients identified as representative of the respective category represented by the classifier;
wherein the health care information, for each patient in the plurality of patients, comprises at least
i. for each category in which the patient is classified, respective patient reported outcome data for the patient for the category, wherein the patient reported outcome data comprises data representing quality of life data reported by the patient,
ii. for each category in which the patient is classified, respective clinical response data for the patient for the category, wherein clinical response data comprises data representing observations made by health care professionals of the patient, and
iii. events of interest data, wherein the events of interest data comprises data representing occurrences of specified negative events in health of the patient;
a data input module, comprising computer program instructions that, when processed by the processing system, configure the processing system to receive and store, in the health care information for the patients in the computer storage, data including patient reported outcome data, clinical response data, and events of interest data for patients; and
an outcome score processing module, comprising computer program instructions that, when processed by the processing system, configure the processing system to be comprising an input connected to the computer storage medium to receive the health care information for the plurality of patients, and an output providing data to the computer storage medium indicative of respective outcome scores for each patient for each of the respective categories in which the patient is classified, the outcome score processing module being configured to, for each patient for each category, compute, and store in the health care information for the patient in the computer storage, the outcome score for the category for the patient using an outcome function defined for the category;
wherein the outcome function for each category is standardized across all patients classified in the category, and wherein the outcome function for each category is based on the clinical response data for the patient for the category, the patient reported outcome data for the patient for the category, and the events of interest data for the patient.
|