Fair ultrasound diagnosis via adversarial protected attribute aware perturbations on latent embeddings.
Authors
Affiliations (9)
Affiliations (9)
- School of Biomedical Engineering, Division of Life Sciences and Medicine, University of Science and Technology of China (USTC), Hefei, Anhui, China.
- Center for Medical Imaging, Robotics, Analytic Computing and Learning (MIRACLE), Suzhou Institute for Advance Research, USTC, Suzhou, Jiangsu, China.
- Key Laboratory of Intelligent Information Processing of Chinese Academy of Sciences (CAS), Institute of Computing Technology, CAS, Beijing, China.
- School of Computer Science and Technology, Harbin Institute of Technology, Harbin, Heilongjiang, China.
- Ultrasound Department, The Affiliated Hospital of Qingdao University, Qingdao, Shangdong, China.
- School of Biomedical Engineering, Division of Life Sciences and Medicine, University of Science and Technology of China (USTC), Hefei, Anhui, China. [email protected].
- Center for Medical Imaging, Robotics, Analytic Computing and Learning (MIRACLE), Suzhou Institute for Advance Research, USTC, Suzhou, Jiangsu, China. [email protected].
- Key Laboratory of Intelligent Information Processing of Chinese Academy of Sciences (CAS), Institute of Computing Technology, CAS, Beijing, China. [email protected].
- State Key Laboratory of Precision and Intelligent Chemistry, University of Science and Technology of China, Hefei, Anhui, China. [email protected].
Abstract
Deep learning techniques have significantly enhanced the convenience and precision of ultrasound image diagnosis, particularly in the crucial step of lesion segmentation. However, recent studies reveal that both train-from-scratch models and pre-trained models often exhibit performance disparities across sex and age attributes, leading to biased diagnoses for different subgroups. In this paper, we propose APPLE, a novel approach designed to mitigate unfairness without altering the parameters of the base model. APPLE achieves this by learning fair perturbations in the latent space through a generative adversarial network. Extensive experiments on both a publicly available dataset and an in-house ultrasound image dataset demonstrate that our method improves segmentation and diagnostic fairness across all sensitive attributes and various backbone architectures compared to the base models. Through this study, we aim to highlight the critical importance of fairness in medical segmentation and contribute to the development of a more equitable healthcare system.