Method and system for predicting adherence to a treatment
AssigneeFair Isaac Corporation
InventorsJun Hua, Hui Zhu, Catherine V. Orate-Pott, David Shellenberger, Deonadayalan Narayanaswamy, Niranjan A. Shetty
Life of this patent
Estimating the term…
Abstract
Data characterizing an individual is received. Thereafter, one or more variables are extracted from the data so that, using a predictive model populated with the extracted variables, a likelihood of the individual adhering to a treatment regimen can be determined. The predictive model is trained on historical treatment regimen adherence data empirically derived from a plurality of subjects. Subsequently, data characterizing the determined likelihood of adherence can be promoted. Related apparatus, systems, techniques and articles are also described.
Claims12
A computer-implemented patient treatment method comprising:
- receiving data characterizing a patient under medical treatment to determine treatment adherence patterns for the patient;
- extracting one or more extracted variables from the data, based on the treatment adherence patterns for the patient, the one or more extracted variables comprising a combination of a plurality of variables including one or more of an income of the patient, a credit risk of the patient, and a geographic region of the patient;
- populating a predictive model with the extracted variables;
- training the predictive model on historical treatment regimen adherence data empirically derived from a plurality of patients previously undergoing medical treatment;
- using the predictive model to determine a likelihood of the patient adhering to the treatment regimen;
- using the predictive model to generate a treatment score indicative of the likelihood of adherence to the treatment regimen by the patient, the treatment score lying within a range of treatment scores;
- using the predictive model to generate a patient response score indicative of likelihood of the patient responding to the treatment regimen;
- based on the treatment score and the patient response score, administering to the patient at least a treatment that yields greater precision in segmentation and agility in treatment actions as applied to the particular patient based on the predictive model's evaluation of the particular patient's characteristic against known characteristics of the plurality of patients previously undergoing medical treatment.
Dependencies are inferred from the claim text.