Facilitating Trust Calibration in Artificial Intelligence–Driven Diagnostic Decision Support Systems for Determining Physicians’ Diagnostic Accuracy: Quasi-Experimental Study
Tetsu Sakamoto, Yukinori Harada, Taro Shimizu, Amaryllis Mavragani · 2024
Read Facilitating Trust Calibration in Artificial Intelligence–Driven Diagnostic Decision Support Systems for Determining Physicians’ Diagnostic Accuracy: Quasi-Experimental Study by Tetsu Sakamoto, Yukinori Harada, Taro Shimizu, Amaryllis Mavragani (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-11612524.txt.
Description
Tetsu Sakamoto and co-authors present a 2024 study of trust in AI diagnostic help among Japanese doctors. AI lists can aid diagnosis, but blind trust in them invites error. Their twist: ask doctors in the test group to judge whether the true diagnosis sits inside the AI's list of ten, a trust calibration step.
Twenty doctors each worked twenty cases drawn from real visits. Diagnostic accuracy was much the same in both groups, near the mid forty percent range. Calibration itself was right only about six times in ten. Yet when calibration was right, doctors did far better. Confidence in AI did not explain the pattern. Better calibration tools may matter more than better lists.
Catalog record
| CREATOR | Tetsu Sakamoto, Yukinori Harada, Taro Shimizu, Amaryllis Mavragani |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 30 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |