Intelligent Deep Feature Fusion for Automated Monkeypox Detection

Abstract: Monkeypox (Mpox), a zoonotic viral disease, has re-emerged as a significant global public health challenge, primarily due to its increasing transmissibility and distinctive dermatolo gical manifestations. Early and accurate identification of the disease is crucial for effective clinical intervention and outbreak prevention. In this study, we propose a hybrid intelligent framework for automated Mpox detection that integrates deep learning-based feature extraction with conventional machine learning algorithms. Specifically, a fine-tuned MobileNetV3 architecture, leveraging transfer learning, is employed to extract representative and discriminative features from skin lesion images. The resulting feature representations are then fed into three machine learning classifiers: Multi-Layer Perceptron (MLP), Random Forest (RF), and Extreme Gradient Boosting (XGBoost). The proposed framework is evaluated on publicly available datasets across both binary and multi class classification scenarios. Experimental findings indicate that the hybrid learning strategies con sistently outperform the standalone MobileNetV3 model. In particular, the MobileNetV3-MLP combination achieves the best performance, attaining an accuracy of 98.05% for multi-class classi fication and 100% for binary classification, compared with 92.68% achieved by the standalone Mo bileNetV3 model. These findings demonstrate the potential of combining deep feature representa tions with traditional machine learning classifiers to enhance the accuracy and reliability of auto mated Mpox detection, supporting the development of intelligent tools for early disease screening. Keywords: Deep Learning, Machine Learning, Medical Image Analysis, Convolutional Neural Networks, Transfer Learning, Mpox Detection, Automated Medical Diagnosis

Authors

Institutions

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-13
DOI
https://doi.org/10.5281/zenodo.24184822
Primary Topic
Poxvirus research and outbreaks
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Intelligent Deep Feature Fusion for Automated Monkeypox Detection

Réda Adjoudj, Ali Kourtiche, Souaad Belhia, Mohammed El Amin Said et al.
2 links
Zenodo (CERN European Organization for Nuclear Research)
Poxvirus research and outbreaks
article

Intelligent Deep Feature Fusion for Automated Monkeypox Detection

Réda Adjoudj, Ali Kourtiche, Souaad Belhia, Mohammed El Amin Said, Mahmoud Fahsi, Anis Allal, Douaa Chikh Benhadjar
article en

Abstract

Abstract: Monkeypox (Mpox), a zoonotic viral disease, has re-emerged as a significant global public health challenge, primarily due to its increasing transmissibility and distinctive dermatolo gical manifestations. Early and accurate identification of the disease is crucial for effective clinical intervention and outbreak prevention. In this study, we propose a hybrid intelligent framework for automated Mpox detection that integrates deep learning-based feature extraction with conventional machine learning algorithms. Specifically, a fine-tuned MobileNetV3 architecture, leveraging transfer learning, is employed to extract representative and discriminative features from skin lesion images. The resulting feature representations are then fed into three machine learning classifiers: Multi-Layer Perceptron (MLP), Random Forest (RF), and Extreme Gradient Boosting (XGBoost). The proposed framework is evaluated on publicly available datasets across both binary and multi class classification scenarios. Experimental findings indicate that the hybrid learning strategies con sistently outperform the standalone MobileNetV3 model. In particular, the MobileNetV3-MLP combination achieves the best performance, attaining an accuracy of 98.05% for multi-class classi fication and 100% for binary classification, compared with 92.68% achieved by the standalone Mo bileNetV3 model. These findings demonstrate the potential of combining deep feature representa tions with traditional machine learning classifiers to enhance the accuracy and reliability of auto mated Mpox detection, supporting the development of intelligent tools for early disease screening. Keywords: Deep Learning, Machine Learning, Medical Image Analysis, Convolutional Neural Networks, Transfer Learning, Mpox Detection, Automated Medical Diagnosis

Zenodo (CERN European Organization for Nuclear Research)
Université Djilali de Sidi Bel Abbès (DZ)
Good health and well-being
Poxvirus research and outbreaks
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon