A novel multi-scale context aggregation and feature pooling network for Mpox classification

Mehdhar S. A. M. Al-Gaashani

doi:10.1016/j.bspc.2025.108254

A novel multi-scale context aggregation and feature pooling network for Mpox classification

10.1016/j.bspc.2025.108254

2025-07-22

0

PRE

AI

Save

Share

Original

0Request

Assist

Abstract

En 中文

• Developed a novel deep learning model integrating Multi-Scale Context Aggregation (MSCA) and Feature Pooling for Mpox skin disease classification. • Conducted extensive evaluations on four diverse datasets and demonstrated strong generalizability through cross-domain experiments on different dataset. • Implemented Grad-CAM and LIME analysis to enhance interpretability, providing insights into the decision-making process for identifying monkeypox lesions. • Proposed model combines multi-scale feature extraction and pooling mechanisms for improved classification, outperforming existing methods. • The model, based on MobileNetV2, maintains computational efficiency, making it suitable for real-world medical applications in low-resource settings.

Keywords:

Mpox classification

deep learning

Multi-Scale Context Aggregation

feature pooling

interpretability

MobileNetV2

A novel multi-scale context aggregation and feature pooling network for Mpox classification

Abstract

Journal

Researchers

Organization

Cited Papers

Citing Papers