When complexity does not pay: benchmarking deep learning and ensemble methods for biomarker discovery
Cyrille Mesue Njume, Irene Petracci, Sonia Bellini, Katarzyna Goljanek-Whysall, Leo R Quinlan, Agnieszka Fiszer, Barbara Borroni, Roberta Ghidoni, Asli Kumbasar, Ali Cakmak · 2026
Read When complexity does not pay: benchmarking deep learning and ensemble methods for biomarker discovery by Cyrille Mesue Njume, Irene Petracci, Sonia Bellini, Katarzyna Goljanek-Whysall, Leo R Quinlan, Agnieszka Fiszer, Barbara Borroni, Roberta Ghidoni, Asli Kumbasar, Ali Cakmak (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-13155123.txt.
Description
Cyrille Mesue Njume and co-authors test whether complex AI methods really help find disease markers, in a 2026 study. Rich multi-layer data sets promise strong markers for disease. Yet ever more complex tools may not pay off in practice. The team built a broad benchmark to check.
They compared 27 ways to select features and 11 prediction models. Three real cohorts covered Alzheimer disease, a related brain disorder, and breast cancer. Group-based feature selection made marker panels more stable and accurate. Deep learning did not beat simple models such as logistic regression. The results argue for care in adding complexity: small, sturdy panels from simpler methods may serve clinics better.
Catalog record
| CREATOR | Cyrille Mesue Njume, Irene Petracci, Sonia Bellini, Katarzyna Goljanek-Whysall, Leo R Quinlan, Agnieszka Fiszer, Barbara Borroni, Roberta Ghidoni, Asli Kumbasar, Ali Cakmak |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 70 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |