EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Deep Learning Framework with Metaheuristic Hyperparameter Optimization for Multi-Class Pox Skin Disease Classification

Noha El-menbawy, Eman El-Gendy, Mohamed Abdelsalam

International Journal of Computational and Experimental Science and Engineering · 25.07.2026 · 12 (3)

DOI: 10.22399/ijcesen.5435 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Computational and Experimental Science and Engineering, online erschienen am 25.07.2026. Verfasst von Noha El-menbawy und 2 weiteren Autor:innen. Die Arbeit nennt 32 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The precise and automated categorization of Monkeypox remains a challenging task due to its visual similarity to other infectious skin diseases, such as Chickenpox and Measles, as well as the limited availability of annotated medical datasets. This study suggests Conv2iT-BCAF, a novel hybrid framework that integrates parallel ConvNeXt and Vision Transformer (viT) branches with Bidirectional cross-attention and Adaptive Fusion to acquire robust local and global representations. Furthermore, an image preprocessing technique, namely Self-Dynamic Hybrid Adaptive CLAHE and Adaptive Gamma correction (SD-HACAG), is developed to enhance the image quality. Moreover, the hyperparameters of Conv2iT-BCAF are optimized using the Black-necked Crane Optimizer (BNCO) metaheuristic algorithm, for finding the optimal model configuration. The experimental results on the MSID dataset show that the proposed Conv2iT-BCAF framework achieves an accuracy of 98.89%, precision of 98.94%, recall of 98.89%, specificity of 99.63%, AUC of 0.9972, and F1-score of 98.89% which is better than the existing approaches, and demonstrates its effectiveness for automated Monkeypox classification.

Zitieren

APA

El-menbawy, N., El-Gendy, E., & Abdelsalam, M. (2026). A Deep Learning Framework with Metaheuristic Hyperparameter Optimization for Multi-Class Pox Skin Disease Classification. International Journal of Computational and Experimental Science and Engineering, 12(3). https://doi.org/10.22399/ijcesen.5435

Vancouver

El-menbawy N, El-Gendy E, Abdelsalam M. A Deep Learning Framework with Metaheuristic Hyperparameter Optimization for Multi-Class Pox Skin Disease Classification. International Journal of Computational and Experimental Science and Engineering. 2026;12(3). doi:10.22399/ijcesen.5435

BibTeX
@article{Elmenbawy2026,
  author = {El-menbawy, Noha and El-Gendy, Eman and Abdelsalam, Mohamed},
  title = {A Deep Learning Framework with Metaheuristic Hyperparameter Optimization for Multi-Class Pox Skin Disease Classification},
  journal = {International Journal of Computational and Experimental Science and Engineering},
  year = {2026},
  volume = {12},
  number = {3},
  doi = {10.22399/ijcesen.5435}
}
Kostenlos starten →