Data Characterization for Reliable AI in Medicine

• Published in: Communications in computer and information science Vol. 1704; p. 3
• Main Authors: Rajaraman, Sivaramakrishnan, Zamzmi, Ghada, Yang, Feng, Xue, Zhiyun, Antani, Sameer K
• Format: Journal Article Book Chapter
• Language: English
• Published: Switzerland 2023
• Subjects: generalizability; deep learning; data-driven design; reliability; robustness; Data characteristics; machine learning; artificial intelligence; medical imaging
• ISSN: 1865-0937; 1865-0929; 1865-0937
• DOI: 10.1007/978-3-031-23599-3_1

Research in Artificial Intelligence (AI)-based medical computer vision algorithms bear promises to improve disease screening, diagnosis, and subsequently patient care. However, these algorithms are highly impacted by the characteristics of the underlying data. In this work, we discuss various data characteristics, namely , and that impact the design, reliability, and evolution of machine learning in medical computer vision. Further, we discuss each characteristic and the recent works conducted in our research lab that informed our understanding of the impact of these characteristics on the design of medical decision-making algorithms and outcome reliability.