US 2023/0142392 A1Published application

Artificial intelligence-based decision management system for appeals on healthcare claim denials

AssigneeVirtusa Corporation

InventorsAlok Mandal, Anirvan Acharya

Life of this patent

Estimating the term…

Loading patent…

Abstract

An artificial intelligence-based decision management system for appeals on healthcare claim denials is disclosed. The plurality of subsystems includes a dataset creation subsystem. The dataset creation subsystem is also configured to determine a plurality of claim predictors corresponding to the one or more context parameters and also configured to create a unique dataset for each of the determined plurality of claim predictors based on values of the one or more context parameters. The plurality of subsystems includes a claim approval propensity prediction subsystem, configured to validate the received appeal by applying the created unique dataset and pre-stored business rules, pre-stored predictive models and pre-stored adaptive models on to a trained artificial intelligence decisioning model and configured to predict claim approval propensity for the appeal based on output of the trained artificial intelligence decisioning model. The data driven decisions clearly increases the correct prediction of claims.

Claims21

Dependent claims

An artificial intelligence-based decision management system for appeals on healthcare claim denials, the artificial intelligence-based decision management system comprising:

a hardware processor;
and
a memory coupled to the hardware processor, wherein the memory comprises a set of program instructions in the form of a plurality of subsystems, configured to be executed by the hardware processor, wherein the plurality of subsystems comprises:
an appeal generation subsystem configured to receive an appeal on healthcare claim from a customer, wherein the appeal comprises one or more context parameters;
a dataset creation subsystem configured to:
determine a plurality of claim predictors corresponding to the one or more context parameters comprised in the received appeal using an artificial intelligence-based predictor model, wherein the plurality of claim predictors comprises a predictor name and predictor type and wherein the artificial intelligence-based predictor model represents the plurality of claim predictors;
and
create a dataset for each of the determined plurality of predictors based on values of the one or more context parameters;
and
a claim approval propensity prediction subsystem configured to:
validate the received appeal by applying the created dataset and pre-stored business rules, pre-stored predictive models and pre-stored adaptive models on to a trained artificial intelligence decisioning model;
and
predict claim approval propensity for the appeal based on output of the trained artificial intelligence decisioning model, wherein the claim approval propensity represents at least one of:
an approval probability score and a rejection probability score for the appeal;
and
an output subsystem configured to output the predicted claim approval propensity for the appeal on a user interface of a user device

13 more elements

A method for operating artificial intelligence-based decision management system for appeals on healthcare claim denials, the method comprising:

receiving, by a processor, an appeal on a healthcare claim from a customer, wherein the appeal comprises one or more context parameters;
determining, by the processor, a plurality of claim predictors corresponding to the one or more context parameters comprised in the received appeal using an artificial intelligence-based predictor model, wherein the plurality of claim predictors comprises a predictor name and predictor type and wherein the artificial intelligence-based predictor model represents the plurality of claim predictors;
creating, by the processor, a dataset for each of the determined plurality of claim predictors based on values of the one or more context parameters;
validating, by the processor, the received appeal by applying the created dataset and pre-stored business rules, pre-stored predictive models and pre-stored adaptive models on to a trained artificial intelligence decisioning model;
predicting, by the processor, claim approval propensity for the appeal based on output of the trained artificial intelligence decisioning model, wherein the claim approval propensity represents at least one of:
an approval probability score and a rejection probability score for the appeal;
and
outputting, by the processor, the predicted claim approval propensity for the appeal on a user interface of a user device.

5 more elements

Dependencies are inferred from the claim text.