Clinical Artificial Intelligence: Design Principles and Fallacies

Matthew B A McDermott; Bret Nestor; Peter Szolovits

doi:10.1016/j.cll.2022.09.004

Clinical Artificial Intelligence: Design Principles and Fallacies

Clin Lab Med. 2023 Mar;43(1):29-46. doi: 10.1016/j.cll.2022.09.004. Epub 2022 Dec 13.

Authors

Matthew B A McDermott¹, Bret Nestor², Peter Szolovits³

Affiliations

¹ CSAIL, MIT, 32 Vassar St, Cambridge, MA 02139, USA. Electronic address: matthew_mcdermott@hms.harvard.edu.
² Department of Computer Science, University of Toronto, 40 St George St, Toronto, ON M5S 2E4, Canada.
³ CSAIL, MIT, 32 Vassar St, Cambridge, MA 02139, USA.

PMID: 36764807
DOI: 10.1016/j.cll.2022.09.004

Abstract

Clinical artificial intelligence (AI)/machine learning (ML) is anticipated to offer new abilities in clinical decision support, diagnostic reasoning, precision medicine, clinical operational support, and clinical research, but careful concern is needed to ensure these technologies work effectively in the clinic. Here, we detail the clinical ML/AI design process, identifying several key questions and detailing several common forms of issues that arise with ML tools, as motivated by real-world examples, such that clinicians and researchers can better anticipate and correct for such issues in their own use of ML/AI techniques.

Keywords: Artificial intelligence; Irresponsibility; Machine learning; Misspecification; Uninterpretability.

Publication types

Review
Research Support, N.I.H., Extramural
Research Support, Non-U.S. Gov't

MeSH terms

Artificial Intelligence*
Decision Support Systems, Clinical*
Machine Learning
Precision Medicine