US 11,990,229 B2
Healthcare provider claims denials prevention systems and methods
Allison Gilmore, Redwood City, CA (US); Tzu-Wei Powers, Mountain View, CA (US); Alan Lehman, Palo Alto, CA (US); and Cindy Zhang, San Jose, CA (US)
Assigned to SymphonyAI Sensa LLC, Palo Alto, CA (US)
Filed by Ayasdi AI LLC, Menlo Park, CA (US)
Filed on Mar. 1, 2018, as Appl. No. 15/909,974.
Claims priority of provisional application 62/465,719, filed on Mar. 1, 2017.
Prior Publication US 2018/0254101 A1, Sep. 6, 2018
Int. Cl. G16H 40/20 (2018.01); G06T 11/20 (2006.01); G16H 15/00 (2018.01)
CPC G16H 40/20 (2018.01) [G06T 11/206 (2013.01); G16H 15/00 (2018.01)] 22 Claims
OG exemplary drawing
 
1. A non-transitory computer readable medium including executable instructions, the instructions being executable by at least one processor to perform a method, the method comprising:
performing a topological data analysis process on multidimensional adjudicated claims data, the performing the topological data analysis process including:
receiving the multidimensional adjudicated claims data, the multidimensional adjudicated claims data including information regarding claim denials for health services, the multidimensional adjudicated claims data further including a first set of claims that have been identified as improperly denied as well as a second set of other claims, the first set of claims having been improperly denied based on at least one incorrect health billing code, each of the at least one incorrect health billing code being for a particular medical procedure, the multidimensional adjudicated claims data further including indications of why a claim of the first set of claims was previously improperly denied, the claims of the first set of claims being previously improperly denied for a variety of different reasons;
receiving at least one metric function and at least one lens function selection;
performing the at least one metric function and the at least one lens function on a set of dimensions of the multidimensional adjudicated claims data to map the claims of the first set of claims, the claims of the second set of claims, and the indications to a reference space, wherein receiving the multidimensional adjudicated claims data including the first set of claims, the second set of claims, and the indications occurs before performing the at least one metric function and the at least one lens function;
generating a cover of overlapping sets of the reference space based on a resolution;
clustering the mapped claims in the reference space using the cover to identify nodes in a topological data analysis graph, each node including one or more claims of the first set of claims as well as one or more claims of the second set of claims as members, each node being connected to another node if they share at least one common claim as members;
generating a topological data analysis graph of the nodes and edges;
identifying groups of nodes in the topological data analysis graph based on claims of the first sets of claims that are members of the nodes;
for each group, identifying differentiating drivers that led to the denial of the claims of the first set of claims in the nodes of that group, the identification of differentiating drivers being based on dimensions of the claims in the nodes of that group, the differentiating drivers being particular to that group relative to other groups in the topological data analysis graph, the identification of differentiating drivers being further based on comparison of claims of the first set of claims with claims of the second set of claims in the nodes of that group, at least one of the claims of the first set of claims in the nodes of that group for a patient different from patients for claims of the second set of claims in the nodes of that group; and
generating a denials application user interface depicting a card for each of at least a subset of the identified groups in the topological data analysis graph that includes the claims of the first set of claims that are the members of the nodes, each card indicating a set of primary statistics based on at least one dimension of each of the claims in the nodes of that group, for each card, the denials application user interface further depicting the differentiating drivers that led to the denial of claims based on the dimensions of the claims in the nodes of that group.