Detecting algorithmic bias in medical-AI models using trees

TL;DR

This paper introduces a novel framework using CART and conformity scores to detect and analyze algorithmic bias in medical-AI models, validated through synthetic and real hospital data to promote fairness in healthcare decision-making.

Contribution

The paper presents an innovative bias detection method specifically tailored for medical-AI systems, employing CART with conformity scores to identify bias regions effectively.

Findings

Successfully identifies bias areas in synthetic data experiments.

Demonstrates practical bias detection in real hospital data.

Validates the approach's effectiveness for clinical fairness.

Abstract

With the growing prevalence of machine learning and artificial intelligence-based medical decision support systems, it is equally important to ensure that these systems provide patient outcomes in a fair and equitable fashion. This paper presents an innovative framework for detecting areas of algorithmic bias in medical-AI decision support systems. Our approach efficiently identifies potential biases in medical-AI models, specifically in the context of sepsis prediction, by employing the Classification and Regression Trees (CART) algorithm with conformity scores. We verify our methodology by conducting a series of synthetic data experiments, showcasing its ability to estimate areas of bias in controlled settings precisely. The effectiveness of the concept is further validated by experiments using electronic medical records from Grady Memorial Hospital in Atlanta, Georgia. These tests demonstrate the practical implementation of our strategy in a clinical environment, where it can function as a vital instrument for guaranteeing fairness and equity in AI-based medical decisions.