US 12,191,010 B2
Systems and methods for machine learning based error correction in large data structures
Kyle Freese, Scottsdale, AZ (US); Sawyer Koops, Phoenix, AZ (US); and Nate Meckes, Durham, NC (US)
Assigned to STChealth, LLC
Filed by STChealth, LLC, Phoenix, AZ (US)
Filed on Jan. 31, 2023, as Appl. No. 18/104,195.
Prior Publication US 2024/0257927 A1, Aug. 1, 2024
Int. Cl. G06F 16/215 (2019.01); G16H 10/60 (2018.01)
CPC G16H 10/60 (2018.01) [G06F 16/215 (2019.01)] 27 Claims
OG exemplary drawing
 
1. A consumer health records access platform including a software application for operation on a user device, the system comprising:
a processor;
a tangible, non-transitory electronic memory in electronic communication with the processor,
a set of computer readable code on the non-transitory electronic memory, including:
a consumer module configured for use by consumers to support registering consumers for a consumer account pending authentication and approval by a healthcare provider;
a provider module configured to enable, via a healthcare provider account, the healthcare provider to look up the consumer account pending authentication and to authenticate and approve or reject creation of the consumer account;
a recording agency module configured for use by recording agency representatives to create, via a recording agency account, the healthcare provider account;
a consumer health record access database comprising a plurality health records associated on a many to one basis with each of a plurality of consumer accounts;
a training data set of health records;
a plurality of factors stored in association with each health record associated with a respective consumer account;
a data processing module, executable by the processor, configured to apply algorithmic transformations to the plurality of health records and comprising a machine learning module including a self learning system, a training system, and a feedback system,
wherein, in response to execution by the processor, the self learning system automatically calculates a first set of weights associated with a first set of factors of a plurality of factors stored in association with each health record of the training data set of health records, and enters the first set of weights and first set of associated factors into a similarity score model;
a similarity threshold value;
a dissimilarity threshold value;
a factor entering module, executable by the processor, to enter at least one factor from the plurality of factors stored in association with each health record associated with the respective consumer account into a similarity score module, wherein the similarity score module is executable by the processor to calculate a similarity score for each one of the plurality of health records associated with the respective customer account, wherein a respective similarity score for the each one of the plurality of health records is based on at least one weight of the first set of weights and at least one factor of the first set of factors of the similarity score model;
a comparator that is executable by the processor to compare the similarity score for each record with the similarity threshold value and the dissimilarity threshold value to determine a subset of the plurality of health records that exceed either value, wherein the comparator allows for comparing the similarity score for each health record of a set of incoming health records received via bi-directional HL7 messaging from each of the healthcare provider and a state immunization registry to the similarity score for each of the plurality of health records of the consumer health record access database;
a flagging unit that is executable by the processor to flag health records of the plurality of health records as similar or as dissimilar based on an output of the comparator; and
a corrective action module that is executable by the processor to perform a corrective action only for those consumer accounts which have associated health records flagged by the flagging unit.