Exploring visual language models as a powerful tool in the diagnosis of Ewing Sarcoma
Álvaro Pastor-Naranjo 1,∗, Pablo Meseguer 1,2, Rocío del Amor 1, Jose Antonio Lopez-Guerrero 3, Samuel Navarro 4,5, Katia Scotlandi 6, Antonio Llombart-Bosch 4, Isidro Machado 4,5,7,8, Valery Naranjo 1 · 2025
Read Exploring visual language models as a powerful tool in the diagnosis of Ewing Sarcoma by Álvaro Pastor-Naranjo 1,∗, Pablo Meseguer 1,2, Rocío del Amor 1, Jose Antonio Lopez-Guerrero 3, Samuel Navarro 4,5, Katia Scotlandi 6, Antonio Llombart-Bosch 4, Isidro Machado 4,5,7,8, Valery Naranjo 1 (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-2501-08042.txt.
Description
Álvaro Pastor-Naranjo and co-authors test vision-language AI on a childhood bone cancer in this 2025 paper. Ewing sarcoma grows as dense sheets of small round cells with little structure, which makes it hard to tell from look-alike tumors. The patients are mostly teens.
The team fed digitized tissue samples to models pre-trained in different ways. One route used plain image training; the other paired images with text, then adapted that pairing to medical data. Within a framework that weighs many small tissue patches, the text-paired route read the cases far more accurately. It also needed far fewer trainable parts and less computing. The authors say vision-language training is a strong base for rare cancer diagnosis.
Catalog record
| CREATOR | Álvaro Pastor-Naranjo 1,∗, Pablo Meseguer 1,2, Rocío del Amor 1, Jose Antonio Lopez-Guerrero 3, Samuel Navarro 4,5, Katia Scotlandi 6, Antonio Llombart-Bosch 4, Isidro Machado 4,5,7,8, Valery Naranjo 1 |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 18 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-08042