US 11,791,038 B1
Machine learning system for asymmetrical matching of care givers and care receivers
Stephen Brobst, Las Vegas, NV (US); Michael Sheehan, Flower Mound, TX (US); John Trustman, Frisco, TX (US); and Michael McDonald, Dallas, TX (US)
Assigned to Zengine Limited, London (GB)
Filed by IntelliCentrics, Inc., Flower Mound, TX (US)
Filed on Nov. 2, 2022, as Appl. No. 17/979,602.
Int. Cl. G16H 40/20 (2018.01); G16H 10/60 (2018.01)
CPC G16H 40/20 (2018.01) [G16H 10/60 (2018.01)] 20 Claims
OG exemplary drawing
 
1. A system for selection of a medical provider (MP) by a requesting patient based on preferences for the MP, comprising:
a first primary model for modeling preferences of a plurality of patients for select MPs and trained on a first training data set comprised of patients and MPs, wherein the patients in the first training data set have rated at least one of the MPs in the first training data set by a predetermined set of patient determined preferences, the first primary model having a first model patient input for receiving information representative of the requesting patient, a first model MP input for receiving information representative of an available MP and a MP preference output for outputting a patient determined preference of the requesting patient for an available MP based on a learned representation of the patient determined preferences in the first training data set in response to a query from the requesting patient;
a second modifying model for modeling preferences of a plurality of MPs for select patients and trained on a second training data set comprised of MPs and patients, wherein the MPs in the second training data set have rated at least one of the patients in the second training data set by a predetermined set of MP determined preferences, the second modifying model having a second model patient input for receiving information representative of the requesting patient, a second model MP input for receiving information representative of an available MP and an MP determined preference output for outputting an MP determined preference of an available MP for the requesting patient based on a learned representation of the MP determined preferences in the second training data set in response to the query to the first primary model from the requesting patient;
an available MP data set containing information representative of available MPs for selection thereof by the requesting patient;
a controller for controlling the first primary model and second modifying model to select at least a portion of the available MPs in the available MP data set for input of the associated information representative of the at least a portion of the available MPs to the first primary model and the second modifying model MP inputs of the first primary model and the second modifying model;
the controller controlling the first primary model and the second modifying model to sequence through the selected at least a portion of the available MPs;
a patient rank calculator to accumulate the output patient determined preferences for MPs associated with the selected at least a portion of the available MPs and generate a MP ranking for each of the selected at least a portion of the available MPs;
a MP rank calculator to accumulate the output MP determined preferences for the requesting patient for each of the selected at least a portion of the available MPs and generate a patient ranking for each of the selected at least a portion of the available MPs; and
a modifier for modifying the patient ranking output of the patient rank calculator with the output of the MP rank calculator to provide as an output in response to the query from the requesting patient a resultant ranking list for each of the selected at least a portion of the available MPs.