Automatic segmentation of knee bony anatomy is essential in orthopedics, and it has been around for several years in both pre-operative and post-operative settings. While deep learning algorithms have demonstrated exceptional performance in medical image analysis, the assessment of fairness and potential biases within these models remains limited. This study aims to revisit deep learning-powered knee-bony anatomy segmentation using plain radiographs to uncover visible gender and racial biases. The current contribution offers the potential to advance our understanding of biases, and it provides practical insights for researchers and practitioners in medical imaging. The proposed mitigation strategies mitigate gender and racial biases, ensuring fair and unbiased segmentation results. Furthermore, this work promotes equal access to accurate diagnoses and treatment outcomes for diverse patient populations, fostering equitable and inclusive healthcare provision.
@article{arxiv.2308.04356,
title = {Learning Unbiased Image Segmentation: A Case Study with Plain Knee Radiographs},
author = {Nickolas Littlefield and Johannes F. Plate and Kurt R. Weiss and Ines Lohse and Avani Chhabra and Ismaeel A. Siddiqui and Zoe Menezes and George Mastorakos and Sakshi Mehul Thakar and Mehrnaz Abedian and Matthew F. Gong and Luke A. Carlson and Hamidreza Moradi and Soheyla Amirian and Ahmad P. Tafti},
journal= {arXiv preprint arXiv:2308.04356},
year = {2023}
}