Drukker Karen, Chen Weijie, Gichoya Judy, Gruszauskas Nicholas, Kalpathy-Cramer Jayashree, Koyejo Sanmi, Myers Kyle, Sá Rui C, Sahiner Berkman, Whitney Heather, Zhang Zi, Giger Maryellen
The University of Chicago, Department of Radiology, Chicago, Illinois, United States.
US Food and Drug Administration, Division of Imaging, Diagnostics, and Software Reliability, Office of Science and Engineering Laboratories, Center for Devices and Radiological Health, Silver Spring, Maryland, United States.
J Med Imaging (Bellingham). 2023 Nov;10(6):061104. doi: 10.1117/1.JMI.10.6.061104. Epub 2023 Apr 26.
To recognize and address various sources of bias essential for algorithmic fairness and trustworthiness and to contribute to a just and equitable deployment of AI in medical imaging, there is an increasing interest in developing medical imaging-based machine learning methods, also known as medical imaging artificial intelligence (AI), for the detection, diagnosis, prognosis, and risk assessment of disease with the goal of clinical implementation. These tools are intended to help improve traditional human decision-making in medical imaging. However, biases introduced in the steps toward clinical deployment may impede their intended function, potentially exacerbating inequities. Specifically, medical imaging AI can propagate or amplify biases introduced in the many steps from model inception to deployment, resulting in a systematic difference in the treatment of different groups.
Our multi-institutional team included medical physicists, medical imaging artificial intelligence/machine learning (AI/ML) researchers, experts in AI/ML bias, statisticians, physicians, and scientists from regulatory bodies. We identified sources of bias in AI/ML, mitigation strategies for these biases, and developed recommendations for best practices in medical imaging AI/ML development.
Five main steps along the roadmap of medical imaging AI/ML were identified: (1) data collection, (2) data preparation and annotation, (3) model development, (4) model evaluation, and (5) model deployment. Within these steps, or bias categories, we identified 29 sources of potential bias, many of which can impact multiple steps, as well as mitigation strategies.
Our findings provide a valuable resource to researchers, clinicians, and the public at large.
为了识别并解决对算法公平性和可信度至关重要的各种偏差来源，并为人工智能在医学成像中的公正和平等部署做出贡献，人们对开发基于医学成像的机器学习方法（也称为医学成像人工智能（AI））以用于疾病的检测、诊断、预后和风险评估并实现临床应用的兴趣与日俱增。这些工具旨在帮助改善医学成像中传统的人类决策。然而，在迈向临床部署的过程中引入的偏差可能会阻碍其预期功能的发挥，从而可能加剧不平等现象。具体而言，医学成像人工智能可能会传播或放大从模型创建到部署的多个步骤中引入的偏差，导致不同群体在治疗上出现系统性差异。
我们的多机构团队包括医学物理学家、医学成像人工智能/机器学习（AI/ML）研究人员、AI/ML偏差专家、统计学家、医生以及来自监管机构的科学家。我们确定了AI/ML中的偏差来源、针对这些偏差的缓解策略，并制定了医学成像AI/ML开发最佳实践的建议。
确定了医学成像AI/ML路线图中的五个主要步骤：（1）数据收集，（2）数据准备和标注，（3）模型开发，（4）模型评估，以及（5）模型部署。在这些步骤或偏差类别中，我们确定了29个潜在偏差来源，其中许多会影响多个步骤，同时还确定了缓解策略。
我们的研究结果为研究人员、临床医生和广大公众提供了宝贵的资源。
