Machine Learning for Medicine Must Be Interpretable, Shareable, Reproducible and Accountable by Design
Ayyüce Begüm Bektaş, Mithat Gönen[0.5em] · 2025
Read Machine Learning for Medicine Must Be Interpretable, Shareable, Reproducible and Accountable by Design by Ayyüce Begüm Bektaş, Mithat Gönen[0.5em] (2025) 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/arxiv-2508-16097.txt.
Description
Ayyüce Begüm Bektaş and Mithat Gönen argue that medical AI must be built for trust from the start, in a 2025 paper. Models that predict survival or risk can be very accurate yet hide how they reach a call. In a clinic, hidden logic blocks trust, review, and legal sign-off. The authors set four design bars: models should be readable, shareable, repeatable, and answerable for their calls.
They point to model styles that show their work, such as sparse kernel methods and case-based learners. Such styles can match black-box scores while letting doctors inspect each call. The paper also urges hard tests for fairness and honest doubt ranges on each output. New tools for shared training and fake-data making can spread work across hospitals without moving patient data. Built this way, the authors say, medical AI can earn a place in real care.
Catalog record
| CREATOR | Ayyüce Begüm Bektaş, Mithat Gönen[0.5em] |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 34 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2508-16097