EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

ARTEMIS: An Explainable AI Framework for Multi-Class COVID-19 Diagnosis with a Newly Curated Dataset

Muhammet Emin Sahin, Hasan Ulutas iD, Mustafa Fatih Erkoc, Baris Karakaya iD, Recep Batuhan Günay iD, Enes Eren Suzgen

Bioengineering · 20.05.2026 · 13 (5) · 588

DOI: 10.3390/bioengineering13050588 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioengineering, online erschienen am 20.05.2026. Verfasst von Muhammet Emin Sahin und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Engineering, Izmir Bakırçay University, Izmir 35665, Türkiye. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In this work, we propose ARTEMIS, a novel and highly interpretable deep learning pipeline for the automatic classification of Chest X-ray (CXR) and Computed Tomography (CT) images into different categories related to important clinical outcomes: COVID-19 infection, Community-Acquired Pneumonia (CAP) cases, and Normal cases. Unlike existing models based on the static feature enhancement step, ARTEMIS proposes a learnable preprocessing component that dynamically adapts the image contrast and sharpness in training mode, facilitating adaptive optimization. Our hybrid network combines EfficientNet-B0 backbone with built-in SE attention with the optional lightweight Transformer encoder block to jointly learn local radiological features and global relationships between pixels. Comprehensive experiments have been conducted on five different datasets, which comprise four publicly available ones and one novel CT dataset annotated by radiologists, including X-ray and CT modalities. Experimental results show strong robustness and generalization with macro F1-scores greater than 96% on public datasets and 99.39% accuracy on our new CT dataset. To interpret the decision-making process, Grad-CAM++ is employed to generate class-discriminative saliency maps; the highlighted regions are systematically validated against established radiological criteria by a board-certified radiologist, confirming that model decisions are grounded in clinically meaningful pulmonary findings rather than imaging artifacts.

Schlagwörter

Stichwörter der Autor:innen

COVID-19CTExplainable AI (Grad-CAM++)X-raydeep learning

Autor:innen und Institutionen

  • Department of Computer Engineering, Izmir Bakırçay University, Izmir 35665, Türkiye.
  • Queen Mary's Digital Environment Research Institute (DERI), London E1 1HH, UK.
  • Department of Computer Engineering, Yozgat Bozok University, Yozgat 66100, Türkiye.
  • Department of Radiology, Faculty of Medicine, Yozgat Bozok University, Yozgat 66100, Türkiye.
  • Department of Electrical Electronics Engineering, Faculty of Engineering, Firat University, Elazig 23119, Türkiye.
  • Department of Computer Technologies, Sorgun Vocational High School, Yozgat Bozok University, Yozgat 66700, Türkiye.

ORCID-Profile: Hasan Ulutas, Baris Karakaya, Recep Batuhan Günay

Zitieren

APA

Sahin, M. E., Ulutas, H., Erkoc, M. F., Karakaya, B., Günay, R. B., & Suzgen, E. E. (2026). ARTEMIS: An Explainable AI Framework for Multi-Class COVID-19 Diagnosis with a Newly Curated Dataset. Bioengineering, 13(5), 588. https://doi.org/10.3390/bioengineering13050588

Vancouver

Sahin ME, Ulutas H, Erkoc MF, Karakaya B, Günay RB, Suzgen EE. ARTEMIS: An Explainable AI Framework for Multi-Class COVID-19 Diagnosis with a Newly Curated Dataset. Bioengineering. 2026;13(5):588. doi:10.3390/bioengineering13050588

BibTeX
@article{Sahin2026,
  author = {Sahin, Muhammet Emin and Ulutas, Hasan and Erkoc, Mustafa Fatih and Karakaya, Baris and Günay, Recep Batuhan and Suzgen, Enes Eren},
  title = {ARTEMIS: An Explainable AI Framework for Multi-Class COVID-19 Diagnosis with a Newly Curated Dataset},
  journal = {Bioengineering},
  year = {2026},
  volume = {13},
  number = {5},
  pages = {588},
  doi = {10.3390/bioengineering13050588}
}
Kostenlos starten →