Data-Core Systems Assigned Patent for Synthesized Invalid Insurance Claims for Training Artificial Intelligence/Machine Learning Model

ALEXANDRIA, Va., April 29 -- Data-Core Systems, Philadelphia, Pennsylvania, has been assigned a patent (No. 11,315,196, initially filed Jan. 12, 2021) developed by five co-inventors for "synthesized invalid insurance claims for training an artificial intelligence / machine learning model." The co-inventors are Anshuman Narayan, Bristol, Pennsylvania, Jishnu Bhattacharyya, Philadelphia, Pennsylvania, Sin-Min Chang, Fairfield, Connecticut, Rathi Dasgupta, Philadelphia, Pennsylvania, and Pradeep K. Banerjee, Philadelphia, Pennsylvania.

The full-text of the patent can be found at https://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.htm&r=1&f=G&l=50&s1=11,315,196.PN.&OS=PN/11,315,196&RS=PN/11,315,196

The abstract states: "Disclosed embodiments include generating invalid insurance claims based on category pairs from a category matrix derived from a set of known valid insurance claims, wherein an individual invalid insurance claim comprises at least one of: (i) a diagnosis code from a first category of a category pair and a procedure code from a second category of the category pair not in the first category or (ii) a procedure code from the first category of the category pair and a diagnosis code from the second category of the category pair not in the first category. Some embodiments also include generating a training dataset comprising (i) valid insurance claims from the set of known valid insurance claims and (ii) the invalid insurance claims, and using the training dataset to train an Artificial Intelligence/Machine Learning Engine to classify insurance claims as valid or invalid."

Written by Jayanti Behera; edited by Biyat Pragnya Rout.