US 11,688,513 B2
Systems and methods for prediction based care recommendations
Tayaru Bayyana, Highlands Ranch, CO (US); and Ankur Kaneria, Cedar Park, TX (US)
Assigned to Cigna Intellectual Property, Inc., Wilmington, DE (US)
Filed by Cigna Intellectual Property, Inc., Wilmington, DE (US)
Filed on Apr. 25, 2022, as Appl. No. 17/728,658.
Application 17/728,658 is a continuation of application No. 17/318,648, filed on May 12, 2021, granted, now 11,315,679.
Prior Publication US 2022/0367041 A1, Nov. 17, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G16H 40/20 (2018.01); G16H 50/20 (2018.01); G16H 10/60 (2018.01); G16H 50/80 (2018.01); G06N 20/00 (2019.01); G16H 70/20 (2018.01); G16H 50/70 (2018.01); G06F 16/25 (2019.01); G06Q 40/08 (2012.01)
CPC G16H 40/20 (2018.01) [G06F 16/254 (2019.01); G06N 20/00 (2019.01); G06Q 40/08 (2013.01); G16H 10/60 (2018.01); G16H 50/20 (2018.01); G16H 50/70 (2018.01); G16H 50/80 (2018.01); G16H 70/20 (2018.01)] 18 Claims
OG exemplary drawing
 
1. A method for providing prediction based healthcare recommendations, the method comprising:
in response to receiving an emergency room services request from a client device associated with a patient,
creating a predictive analytics model configured to provide a score indicating a likelihood that the emergency room services request is associated with an avoidable visit, the emergency room services request including a requestor location, a requestor identifier, and requestor symptoms;
training the one or more predictive models using first data collected from health records over a first time period, to create trained models;
validating the trained models using second data collected from the health records over a second time period, to create validated models, the first time period comprising a duration longer than the second time period, and the first time period comprising an age older than the second time period;
determining that supplemental information is needed to predict whether an emergency room visit is avoidable, using the predictive analytics model; and
in response to supplemental information being needed,
transmitting a prompt to the client device associated with a patient, the prompt including a supplemental question;
receiving a response to the supplemental question, from the client device;
designating the response as a feature;
weighting the feature with an associated weight;
applying the predictive analytics model to the feature with the associated weight to score the likelihood that the emergency room visit is avoidable and create a scored likelihood;
generating a prediction based on the predictive analytics model and the scored likelihood; and
routing a request to an alternative services provider or an emergency room services provider, based on the prediction.