Method of differentiating patterns of care

Patent metadata

Field	Value
Publication number	US-2025022613-A1
Application number	US-202318352390-A
Country	US
Kind code	A1
Filing date	Jul 14, 2023
Priority date	Jul 14, 2023
Publication date	Jan 16, 2025
Grant date	—

Abstract

A computer-implemented method for differentiating patterns of care (DPoC) to detect anomalous subsets in any given population with a defined set of outcomes and features. The computer-implemented method includes detecting the anomalous subsets, ranking the anomalous subsets based on a score of each anomalous subset that is reflective of an anomaly thereof, specifying whether each of the anomalous subsets overlaps with another one of the anomalous subsets, whether each of the anomalous subsets is unique and whether each of the anomalous subsets is conditional and specifying as to whether the detecting of each of the anomalous subsets has a higher or lower outcome than expected.

First claim

What is claimed is:

1 . A computer-implemented method for differentiating patterns of care (DPoC) to detect anomalous subsets in any given population with a defined set of outcomes and features, the computer-implemented method comprising:
detecting the anomalous subsets; ranking the anomalous subsets based on a score of each anomalous subset that is reflective of an anomaly thereof; specifying whether each of the anomalous subsets overlaps with another one of the anomalous subsets, whether each of the anomalous subsets is unique and whether each of the anomalous subsets is conditional; and specifying as to whether the detecting of each of the anomalous subsets has a higher or lower outcome than expected.

2 . The computer-implemented method according to claim 1 , wherein the computer-implemented method is characterized in that input data preparation, setup configuration, algorithm executions and report generations are automated.

3 . The computer-implemented method according to claim 2 , wherein the input data comprises binary or numerical outcome data and feature data with an unlimited number of features.

4 . The computer-implemented method according to claim 3 , further comprising a feature selection operation in which the features of the unlimited number of features are selected.

5 . The computer-implemented method according to claim 3 , further comprising a parameter choice operation in which parameters comprise regularization parameters, counts for bootstrap repetitions, randomization initializations, numbers of the anomalous subsets to be identified and anomaly detection directions.

6 . The computer-implemented method according to claim 1 , wherein the specifying of whether each of the anomalous subsets overlaps with another one of the anomalous subsets, whether each of the anomalous subsets is unique and whether each of the anomalous subsets is conditional comprises setting the specifying to specify one of whether each of the anomalous subsets overlaps with another one of the anomalous subsets, whether each of the anomalous subsets is unique or whether each of the anomalous subsets is conditional.

7 . The computer-implemented method according to claim 1 , further comprising summarizing and visualizing results of the detecting, the ranking, the specifying of whether each of the anomalous subsets overlaps with another one of the anomalous subsets, whether each of the anomalous subsets is unique and whether each of the anomalous subsets is conditional and the specifying as to whether the detecting of each of the anomalous subsets has a higher or lower outcome than expected.

8 . A computer program product for differentiating patterns of care (DPoC) to detect anomalous subsets in any given population with a defined set of outcomes and features, the computer program product comprising one or more computer readable storage media having computer readable program code collectively stored on the one or more computer readable storage media, the computer readable program code being executed by a processor of a computer system to cause the computer system to perform a method comprising:
detecting the anomalous subsets; ranking the anomalous subsets based on a score of each anomalous subset that is reflective of an anomaly thereof; specifying whether each of the anomalous subsets overlaps with another one of the anomalous subsets, whether each of the anomalous subsets is unique and whether each of the anomalous subsets is conditional; and specifying as to whether the detecting of each of the anomalous subsets has a higher or lower outcome than expected.

9 . The computer program product according to claim 8 , wherein the method is characterized in that input data preparation, setup configuration, algorithm executions and report generations are automated.

10 . The computer program product according to claim 9 , wherein the input data comprises binary or numerical outcome data and feature data with an unlimited number of features.

11 . The computer program product according to claim 10 , wherein the method further comprises a feature selection operation in which the features of the unlimited number of features are selected.

12 . The computer program product according to claim 10 , wherein the method further comprises a parameter choice operation in which parameters comprise regularization parameters, counts for bootstrap repetitions, randomization initializations, numbers of the anomalous subsets to be identified and anomaly detection directions.

13 . The computer program product according to claim 8 , wherein the specifying of whether each of the anomalous subsets overlaps with another one of the anomalous subsets, whether each of the anomalous subsets is unique and whether each of the anomalous subsets is conditional comprises setting the specifying to specify one of whether each of the anomalous subsets overlaps with another one of the anomalous subsets, whether each of the anomalous subsets is unique or whether each of the anomalous subsets is conditional.

14 . The computer program product according to claim 8 , wherein the method further comprises summarizing and visualizing results of the detecting, the ranking, the specifying of whether each of the anomalous subsets overlaps with another one of the anomalous subsets, whether each of the anomalous subsets is unique and whether each of the anomalous subsets is conditional and the specifying as to whether the detecting of each of the anomalous subsets has a higher or lower outcome than expected.

15 . A computing system comprising:
a processor; a memory coupled to the processor; and one or more computer readable storage media coupled to the processor, the one or more computer readable storage media collectively containing instructions that are executed by the processor via the memory to implement a method for differentiating patterns of care (DPoC) to detect anomalous subsets in any given population with a defined set of outcomes and features comprising: detecting the anomalous subsets; ranking the anomalous subsets based on a score of each anomalous subset that is reflective of an anomaly thereof; specifying whether each of the anomalous subsets overlaps with another one of the anomalous subsets, whether each of the anomalous subsets is unique and whether each of the anomalous subsets is conditional; and specifying as to whether the detecting of each of the anomalous subsets has a higher or lower outcome than expected.

16 . The computing system according to claim 15 , wherein the differentiating of patterns of care (DPoC) to detect anomalous subsets in any given population with the defined set of outcomes and features is characterized in that input data preparation, setup configuration, algorithm executions and report generations are automated.

17 . The computing system according to claim 16 , wherein:
the input data comprises binary or numerical outcome data and feature data with an unlimited number of features, and the instructions further comprise a feature selection operation in which the features of the unlimited number of features are selected.

18 . The computing system according to claim 16 , wherein:
the input data comprises binary or numerical outcome data and feature data with an unlimited number of features, and the instructions further comprise a parameter choice operation in which parameters comprise regularization parameters, counts for bootstrap repetitions, randomization initializations, numbers of the anomalous subsets to be identified and anomaly dete

Assignees

• IBM

Inventors

• Sirbu George
• Ogallo William
• Tadesse Girmaw Abebe
• Xin Feng
• Deshpande Ajay Ashok
• Wachira Charles Muchiri
• Walcott Aisha
• Anand Vibha
• Hodor Brandi
• Savliwala Zainab
• Onando Isaiah Mulang'

Classifications

• G16H50/70 Primary

  for mining of medical data, e.g. analysing previous cases of other patients · CPC title