Detecting Dataset Bias in Medical AI: A Generalized and Modality-Agnostic Auditing Framework

• URL: http://arxiv.org/abs/2503.09969v1
• Date: Thu, 13 Mar 2025 02:16:48 GMT
• Title: Detecting Dataset Bias in Medical AI: A Generalized and Modality-Agnostic Auditing Framework
• Authors: Nathan Drenkow, Mitchell Pavlak, Keith Harrigian, Ayah Zirikly, Adarsh Subbaswamy, Mathias Unberath,
• Abstract summary: latent bias in machine learning datasets can be amplified during training and/or hidden during testing.<n>We present a data modality-agnostic auditing framework for generating targeted hypotheses about sources of bias.<n>We demonstrate the broad applicability and value of our method by analyzing large-scale medical datasets.
• Score: 8.520644988801243
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Data-driven AI is establishing itself at the center of evidence-based medicine. However, reports of shortcomings and unexpected behavior are growing due to AI's reliance on association-based learning. A major reason for this behavior: latent bias in machine learning datasets can be amplified during training and/or hidden during testing. We present a data modality-agnostic auditing framework for generating targeted hypotheses about sources of bias which we refer to as Generalized Attribute Utility and Detectability-Induced bias Testing (G-AUDIT) for datasets. Our method examines the relationship between task-level annotations and data properties including protected attributes (e.g., race, age, sex) and environment and acquisition characteristics (e.g., clinical site, imaging protocols). G-AUDIT automatically quantifies the extent to which the observed data attributes may enable shortcut learning, or in the case of testing data, hide predictions made based on spurious associations. We demonstrate the broad applicability and value of our method by analyzing large-scale medical datasets for three distinct modalities and learning tasks: skin lesion classification in images, stigmatizing language classification in Electronic Health Records (EHR), and mortality prediction for ICU tabular data. In each setting, G-AUDIT successfully identifies subtle biases commonly overlooked by traditional qualitative methods that focus primarily on social and ethical objectives, underscoring its practical value in exposing dataset-level risks and supporting the downstream development of reliable AI systems. Our method paves the way for achieving deeper understanding of machine learning datasets throughout the AI development life-cycle from initial prototyping all the way to regulation, and creates opportunities to reduce model bias, enabling safer and more trustworthy AI systems.

Related papers

• AI Alignment in Medical Imaging: Unveiling Hidden Biases Through Counterfactual Analysis [16.21270312974956]
  We introduce a novel statistical framework to evaluate the dependency of medical imaging ML models on sensitive attributes, such as demographics. We present a practical algorithm that combines conditional latent diffusion models with statistical hypothesis testing to identify and quantify such biases.
  arXiv  Detail & Related papers  (2025-04-28T09:28:25Z)
• Explainable AI for Classifying UTI Risk Groups Using a Real-World Linked EHR and Pathology Lab Dataset [0.47517735516852333]
  We leverage a linked EHR dataset to characterize urinary tract infections (UTIs)<n>We introduce a UTI risk estimation framework informed by clinical expertise to estimate UTI risk across individual patient timelines.<n>Our findings reveal differences in clinical and demographic predictors across risk groups.
  arXiv  Detail & Related papers  (2024-11-26T18:10:51Z)
• Prospector Heads: Generalized Feature Attribution for Large Models & Data [82.02696069543454]
  We introduce prospector heads, an efficient and interpretable alternative to explanation-based attribution methods. We demonstrate how prospector heads enable improved interpretation and discovery of class-specific patterns in input data.
  arXiv  Detail & Related papers  (2024-02-18T23:01:28Z)
• SUDO: a framework for evaluating clinical artificial intelligence systems without ground-truth annotations [3.7525007896336944]
  We introduce SUDO, a framework for evaluating AI systems without ground-truth annotations. We show that SUDO can be a reliable proxy for model performance and thus identify unreliable predictions.
  arXiv  Detail & Related papers  (2024-01-02T18:12:03Z)
• Evaluating the Fairness of the MIMIC-IV Dataset and a Baseline Algorithm: Application to the ICU Length of Stay Prediction [65.268245109828]
  This paper uses the MIMIC-IV dataset to examine the fairness and bias in an XGBoost binary classification model predicting the ICU length of stay. The research reveals class imbalances in the dataset across demographic attributes and employs data preprocessing and feature extraction. The paper concludes with recommendations for fairness-aware machine learning techniques for mitigating biases and the need for collaborative efforts among healthcare professionals and data scientists.
  arXiv  Detail & Related papers  (2023-12-31T16:01:48Z)
• An AI-Guided Data Centric Strategy to Detect and Mitigate Biases in Healthcare Datasets [32.25265709333831]
  We generate a data-centric, model-agnostic, task-agnostic approach to evaluate dataset bias by investigating the relationship between how easily different groups are learned at small sample sizes (AEquity) We then apply a systematic analysis of AEq values across subpopulations to identify and manifestations of racial bias in two known cases in healthcare. AEq is a novel and broadly applicable metric that can be applied to advance equity by diagnosing and remediating bias in healthcare datasets.
  arXiv  Detail & Related papers  (2023-11-06T17:08:41Z)
• Data AUDIT: Identifying Attribute Utility- and Detectability-Induced Bias in Task Models [8.420252576694583]
  We present a first technique for the rigorous, quantitative screening of medical image datasets. Our method decomposes the risks associated with dataset attributes in terms of their detectability and utility. Using our method, we show our screening method reliably identifies nearly imperceptible bias-inducing artifacts.
  arXiv  Detail & Related papers  (2023-04-06T16:50:15Z)
• D-BIAS: A Causality-Based Human-in-the-Loop System for Tackling Algorithmic Bias [57.87117733071416]
  We propose D-BIAS, a visual interactive tool that embodies human-in-the-loop AI approach for auditing and mitigating social biases. A user can detect the presence of bias against a group by identifying unfair causal relationships in the causal network. For each interaction, say weakening/deleting a biased causal edge, the system uses a novel method to simulate a new (debiased) dataset.
  arXiv  Detail & Related papers  (2022-08-10T03:41:48Z)
• TRAPDOOR: Repurposing backdoors to detect dataset bias in machine learning-based genomic analysis [15.483078145498085]
  Under-representation of groups in datasets can lead to inaccurate predictions for certain groups, which can exacerbate systemic discrimination issues. We propose TRAPDOOR, a methodology for identification of biased datasets by repurposing a technique that has been mostly proposed for nefarious purposes: Neural network backdoors. Using a real-world cancer dataset, we analyze the dataset with the bias that already existed towards white individuals and also introduced biases in datasets artificially.
  arXiv  Detail & Related papers  (2021-08-14T17:02:02Z)
• Bootstrapping Your Own Positive Sample: Contrastive Learning With Electronic Health Record Data [62.29031007761901]
  This paper proposes a novel contrastive regularized clinical classification model. We introduce two unique positive sampling strategies specifically tailored for EHR data. Our framework yields highly competitive experimental results in predicting the mortality risk on real-world COVID-19 EHR data.
  arXiv  Detail & Related papers  (2021-04-07T06:02:04Z)
• Estimating and Improving Fairness with Adversarial Learning [65.99330614802388]
  We propose an adversarial multi-task training strategy to simultaneously mitigate and detect bias in the deep learning-based medical image analysis system. Specifically, we propose to add a discrimination module against bias and a critical module that predicts unfairness within the base classification model. We evaluate our framework on a large-scale public-available skin lesion dataset.
  arXiv  Detail & Related papers  (2021-03-07T03:10:32Z)
• Uncovering the structure of clinical EEG signals with self-supervised learning [64.4754948595556]
  Supervised learning paradigms are often limited by the amount of labeled data that is available. This phenomenon is particularly problematic in clinically-relevant data, such as electroencephalography (EEG) By extracting information from unlabeled data, it might be possible to reach competitive performance with deep neural networks.
  arXiv  Detail & Related papers  (2020-07-31T14:34:47Z)
• Causal Feature Selection for Algorithmic Fairness [61.767399505764736]
  We consider fairness in the integration component of data management. We propose an approach to identify a sub-collection of features that ensure the fairness of the dataset.
  arXiv  Detail & Related papers  (2020-06-10T20:20:10Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.