Understanding Bias in Artificial Intelligence: A Practice Perspective

Review Radiology, Nuclear Medicine & Medical Imaging

Imaging Artificial Intelligence: A Framework for Radiologists to Address Health Equity, From the AJR Special Series on DEI

Melissa A. Davis et al.

Summary: Artificial intelligence has great potential in healthcare, particularly in radiology. It can improve patient care pathways and increase efficiency for healthcare practitioners. Additionally, AI holds promise for reducing health disparities and promoting health equity.

AMERICAN JOURNAL OF ROENTGENOLOGY (2023)

添加到收藏夹

Review Clinical Neurology

Ethical Considerations and Fairness in the Use of Artificial Intelligence for Neuroradiology

C. G. Filippi et al.

Summary: This review qualitatively and mathematically defines the concepts of algorithmic bias and fairness, and provides illustrative examples. The importance of explainability, accountability, and transparency in artificial intelligence algorithm development and clinical deployment is discussed, along with steps to mitigate bias and unfairness. The implementation of fairness criteria to maximize benefit and minimize harm to neuroradiology patients is also explored, with suggestions for neuroradiologists as artificial intelligence algorithms become integrated into routine clinical workflow.

AMERICAN JOURNAL OF NEURORADIOLOGY (2023)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Mitigating Bias in Radiology Machine Learning: 1. Data Handling

Pouria Rouzrokh et al.

Summary: Minimizing bias is crucial for the adoption and implementation of machine learning in clinical practice. This report presents 12 suboptimal practices during data handling of an ML study, provides examples, and describes preventive measures.

RADIOLOGY-ARTIFICIAL INTELLIGENCE (2022)

添加到收藏夹

Article Radiology, Nuclear Medicine & Medical Imaging

Diagnostic Accuracy and Failure Mode Analysis of a Deep Learning Algorithm for the Detection of Intracranial Hemorrhage

Andrew F. Voter et al.

Summary: This study revealed decreased diagnostic accuracy of an AI DSS at our institution. Despite extensive evaluation, we were unable to identify the source of this discrepancy, raising concerns about the generalizability of these tools with indeterminate failure modes. These results further highlight the need for standardized study design to allow for rigorous and reproducible site-to-site comparison of emerging deep learning technologies.

JOURNAL OF THE AMERICAN COLLEGE OF RADIOLOGY (2021)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Active Reprioritization of the Reading Worklist Using Artificial Intelligence Has a Beneficial Effect on the Turnaround Time for Interpretation of Head CT with Intracranial Hemorrhage

Thomas J. O'Neill et al.

Summary: This study aimed to optimize the delivery of machine learning techniques in a clinical setting to detect intracranial hemorrhage on non-contrast enhanced CT images. Results showed that introducing an algorithm with flagging feature can reduce wait time, especially for low-priority examinations.

RADIOLOGY-ARTIFICIAL INTELLIGENCE (2021)

添加到收藏夹

Review Radiology, Nuclear Medicine & Medical Imaging