MEL-IA: An Interoperable AI System for Multimodal Skin Lesion Classification in Hospital Settings
Pablo Candela Córcoles, Alberto De Ramón Fernández, Marcelo Saval Calvo, Jose María Salinas Serrano, Diego Guijarro Peral, Daniel Ruiz Fernández · 2026
Read MEL-IA: An Interoperable AI System for Multimodal Skin Lesion Classification in Hospital Settings by Pablo Candela Córcoles, Alberto De Ramón Fernández, Marcelo Saval Calvo, Jose María Salinas Serrano, Diego Guijarro Peral, Daniel Ruiz Fernández (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/pmc-13260036.txt.
Description
Pablo Candela Córcoles and co-authors deploy MEL-IA, an AI system that classifies skin lesions inside a real hospital's information flow. Skin cancer triage needs speed, but many models die at the hospital door for want of integration. MEL-IA fuses dermatoscope images with structured clinical data in one multimodal network.
Their 2026 study trains on large public lesion collections, checks robustness with five-fold validation, and tests generalization on an independent data set. Accuracy stays high across both, with melanoma sensitivity the watched weak point. The decisive test is operational: the system runs live against the hospital's imaging and records standards, exchanging studies and reports in production. Interoperability, the authors argue, is a clinical feature, not an afterthought.
Catalog record
| CREATOR | Pablo Candela Córcoles, Alberto De Ramón Fernández, Marcelo Saval Calvo, Jose María Salinas Serrano, Diego Guijarro Peral, Daniel Ruiz Fernández |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 48 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |