Hybrid ConvNeXtV2–ViT Architecture with Ontology-Driven Explainability and Out-of-Distribution Awareness for Transparent Chest X-Ray Diagnosis

<b>Background:</b> Chest X-ray (CXR) is widely used for the assessment of thoracic diseases, yet automated multi-label interpretation remains challenging due to subtle visual patterns, overlapping anatomical structures, and frequent co-occurrence of abnormalities. While recent deep learning models h...

Popoln opis

Bibliografske podrobnosti

Principais autores:	Naif Almughamisi, Gibrael Abosamra, Adnan Albar, Mostafa Saleh
Format:	Artigo
Jezik:	Inglês
Izdano:	MDPI AG 2026-01-01
Serija:	Diagnostics
Teme:	hybrid CNN–ViT interpretable model grad-CAM neuro-symbolic reasoning thoracic diseases out-of-distribution detection
Online dostop:	https://www.mdpi.com/2075-4418/16/2/294
Oznake:	Označite Brez oznak, prvi označite!