An Attention-Enhanced ViT-HLNN Hybrid Ensemble Framework for Multi-Class Gastrointestinal Disease Classification
Abdullah, Muhammad Ateeb Ather, Zulaikha Fatima, José Luis Oropeza Rodríguez, Carlos Guzmán Sánchez-Mejorada, Rolando Quintero Téllez, Miguel Jesús Torres Ruiz · 2026
Read An Attention-Enhanced ViT-HLNN Hybrid Ensemble Framework for Multi-Class Gastrointestinal Disease Classification by Abdullah, Muhammad Ateeb Ather, Zulaikha Fatima, José Luis Oropeza Rodríguez, Carlos Guzmán Sánchez-Mejorada, Rolando Quintero Téllez, Miguel Jesús Torres Ruiz (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/PMC13346603.txt.
Description
Abdullah and co-authors propose a hybrid deep learning system, issued in 2026, for sorting gut diseases from endoscopy images. Their design joins a Vision Transformer with a layered memory network and then adds bagging and stacking ensembles to combine several model outputs.
Training used the Kvasir-v2 set of over 8,000 labelled images across eight disease classes. The held-out test reached up to 99.99 percent accuracy in the reported runs, with a kappa value of 0.994 that signals very strong agreement. The system beat five named baseline models and stayed above 96 percent correct under added noise, blur, and light shifts, though such near-perfect scores merit care in new clinics.
Catalog record
| CREATOR | Abdullah, Muhammad Ateeb Ather, Zulaikha Fatima, José Luis Oropeza Rodríguez, Carlos Guzmán Sánchez-Mejorada, Rolando Quintero Téllez, Miguel Jesús Torres Ruiz |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 162707 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |