Safety design guidelines for clinician–AI interaction in computer-aided diagnosis systems using system-theoretic framework with explainability validation
Yuki Hagiwara, Katherine Fitch, Mario Trapp · 2026
Read Safety design guidelines for clinician–AI interaction in computer-aided diagnosis systems using system-theoretic framework with explainability validation by Yuki Hagiwara, Katherine Fitch, Mario Trapp (2026) 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/PMC13421477.txt.
Description
Hagiwara and colleagues present a 2026 paper on making computer-aided diagnosis tools safer when doctors work beside artificial intelligence. The authors use a systems method, STPA, which traces how control actions in a work process can turn unsafe, and add checks on whether system explanations hold up.
The analysis flagged hazards such as automation bias, where doctors accept a system answer too readily, and misread explanations that bend judgment without being noticed. From each unsafe action the team derived design guidelines that builders can trace back to a named hazard. The guidelines push for clear transparency, explanations that stay coherent, and trust that matches the tool's real skill. The result is a practical bridge between safety analysis and product design for diagnosis tools, where a wrong call can cause serious harm to patients.
Catalog record
| CREATOR | Yuki Hagiwara, Katherine Fitch, Mario Trapp |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 71 KB |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |