Fairness Metrics in AI Healthcare Applications: A Review

Ibomoiye Domor Mienye, Theo G. Swart, George Obaido

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

As artificial intelligence (AI) systems increasingly become popular in the healthcare sector, it is important to ensure the output of these technologies is fair and bias-free. This paper provides a concise survey of fairness metrics applied in healthcare AI, including their mathematical representations, suitable use cases, and limitations, which are lacking in the existing literature. The study also highlights the significance of implementing fairness metrics to ensure equitable outcomes across diverse patient populations and discusses the challenges and future directions in this rapidly evolving field.

Original languageEnglish
Title of host publicationProceedings - 2024 IEEE International Conference on Information Reuse and Integration for Data Science, IRI 2024
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages284-289
Number of pages6
ISBN (Electronic)9798350351187
DOIs
Publication statusPublished - 2024
Event25th IEEE International Conference on Information Reuse and Integration for Data Science, IRI 2024 - San Jose, United States
Duration: 7 Aug 20249 Aug 2024

Publication series

NameProceedings - 2024 IEEE International Conference on Information Reuse and Integration for Data Science, IRI 2024

Conference

Conference25th IEEE International Conference on Information Reuse and Integration for Data Science, IRI 2024
Country/TerritoryUnited States
CitySan Jose
Period7/08/249/08/24

Keywords

  • AI
  • bias
  • fairness metrics
  • healthcare
  • machine learning

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Information Systems
  • Information Systems and Management
  • Safety, Risk, Reliability and Quality

Fingerprint

Dive into the research topics of 'Fairness Metrics in AI Healthcare Applications: A Review'. Together they form a unique fingerprint.

Cite this