No Masks Needed: Explainable AI for Deriving Segmentation from Classification
Mosong Ma 1 and Tania Stathaki 1 and Michalis Lazarou 2 · 2025
Read No Masks Needed: Explainable AI for Deriving Segmentation from Classification by Mosong Ma 1 and Tania Stathaki 1 and Michalis Lazarou 2 (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-2508-04534.txt.
Description
Mosong Ma, Tania Stathaki, and Michalis Lazarou tackle medical image outlining in a 2025 paper that needs no hand-drawn masks. Outlining organs or flaws in scans is a key step for computer-aided care. Past mask-free methods worked on plain photos but fell short on medical images.
Their method fine-tunes trained models for medical use and adds scores that show which image parts drove a choice. Those scores guide the outlining step itself. Tests on three data sets, covering breast scans, cell images, and gut views, gave better results than older routes. The work points to cheaper and faster scan review in clinics.
Catalog record
| CREATOR | Mosong Ma 1 and Tania Stathaki 1 and Michalis Lazarou 2 |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 29 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2508-04534