Debiased Noise Editing on Foundation Models for Fair Medical Image Classification
Not on the map. Why?
Abstract (arXiv)
In the era of Foundation Models' (FMs) rising prominence in AI, our study addresses the challenge of biases in medical images while the model operates in black-box (e.g., using FM API), particularly spurious correlations between pixels and sensitive attributes. Traditional methods for bias mitigation face limitations due to the restricted access to web-hosted FMs and difficulties in addressing the underlying bias encoded within the FM API. We propose a D(ebiased) N(oise) E(diting) strategy, termed DNE, which generates DNE noise to mask such spurious correlation. DNE is capable of mitigating bias both within the FM API embedding and the images themselves. Furthermore, DNE is suitable for both white-box and black-box FM APIs, where we introduced G(reedy) (Z)eroth-O(rder) (GeZO) optimization for it when the gradient is inaccessible in black-box APIs. Our whole pipeline enables fairness-aware image editing that can be applied across various medical contexts without requiring direct model manipulation or significant computational resources. Our empirical results demonstrate the method's effectiveness in maintaining fairness and utility across different patient groups and diseases. In the era of AI-driven medicine, this work contributes to making healthcare diagnostics more equitable, showcasing a practical solution for bias mitigation in pre-trained image FMs. Our code is provided at https://github.com/ubc-tea/DNE-foundation-model-fairness.
Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 19 of its 30 references that are in the map. First posted on arXiv 10 Mar 2024.
Builds on
19 of 30 references.
References · 19
- 2016Equality of Opportunity in Supervised Learningcited by 2,000
- 2019CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparisoncited by 1,410
- 2019Distributionally Robust Neural Networks for Group Shifts: On the Importance of Regularization for Worst-Case Generalizationcited by 1,232
- 2017ZOO: Zeroth Order Optimization based Black-box Attacks to Deep Neural Networks without Training Substitute Modelscited by 656
- 2023Towards Expert-Level Medical Question Answering with Large Language Modelscited by 382
- 2023A Comprehensive Survey on Pretrained Foundation Models: A History from BERT to ChatGPTcited by 364
- 2023Fine-Tuning Language Models with Just Forward Passescited by 337
- 2021Large Pre-trained Language Models Contain Human-like Biases of What is Right and Wrong to Docited by 178
- 2020A Primer on Zeroth-Order Optimization in Signal Processing and Machine Learningcited by 160
- 2018Achieving Fairness through Adversarial Learning: an Application to Recidivism Predictioncited by 77
- 2022Delving into Masked Autoencoders for Multi-Label Thorax Disease Classificationcited by 65
- 2022FairPrune: Achieving Fairness Through Pruning for Dermatological Disease Diagnosiscited by 59
7 more
- 2021Fairness in Cardiac MR Image Analysis: An Investigation of Bias Due to Data Imbalance in Deep Learning Based Segmentationcited by 57
- 2020Detect and Correct Bias in Multi-Site Neuroimaging Datasetscited by 44
- 2023The Emergence of Essential Sparsity in Large Pre-trained Models: The Weights that Mattercited by 42
- 2023On Fairness of Medical Image Classification with Multiple Sensitive Attributes via Learning Orthogonal Representationscited by 17
- 2023FedSoup: Improving Generalization and Personalization in Federated Learning via Selective Model Interpolationcited by 14
- 2022Improving Fairness in Image Classification via Sketchingcited by 11
- 2021Algorithmic encoding of protected characteristics in image-based models for disease detectioncited by 7
Built on by
Citing papers per year: 2025: 2; 2026: 1.
3 papers.