An Explainable Nature-Inspired Framework for Monkeypox Diagnosis: Xception Features Combined with NGBoost and African Vultures Optimization Algorithm

TL;DR

The paper tackles automated monkeypox diagnosis from skin lesion images by integrating Xception-based deep features, PCA-driven dimensionality reduction, and NGBoost classification, with hyperparameter optimization via the African Vultures Optimization Algorithm (AVOA) and interpretability through Grad-CAM and LIME. On the Monkeypox Skin Lesion Dataset (MSLD), the proposed AVOA-NGBoost framework delivers state-of-the-art performance (accuracy 97.53%, F1 97.72%, AUC 97.47%), while maintaining robust generalization across 5-fold cross-validation. The work highlights the value of probabilistic predictions and uncertainty quantification in medical imaging, coupled with explicit explanations to support clinical trust. The integrated approach offers a scalable, efficient diagnostic tool for early monkeypox detection, particularly in low-resource settings, and points to deployment in mobile or point-of-care contexts where rapid, interpretable decisions are critical.

Abstract

The recent global spread of monkeypox, particularly in regions where it has not historically been prevalent, has raised significant public health concerns. Early and accurate diagnosis is critical for effective disease management and control. In response, this study proposes a novel deep learning-based framework for the automated detection of monkeypox from skin lesion images, leveraging the power of transfer learning, dimensionality reduction, and advanced machine learning techniques. We utilize the newly developed Monkeypox Skin Lesion Dataset (MSLD), which includes images of monkeypox, chickenpox, and measles, to train and evaluate our models. The proposed framework employs the Xception architecture for deep feature extraction, followed by Principal Component Analysis (PCA) for dimensionality reduction, and the Natural Gradient Boosting (NGBoost) algorithm for classification. To optimize the model's performance and generalization, we introduce the African Vultures Optimization Algorithm (AVOA) for hyperparameter tuning, ensuring efficient exploration of the parameter space. Our results demonstrate that the proposed AVOA-NGBoost model achieves state-of-the-art performance, with an accuracy of 97.53%, F1-score of 97.72% and an AUC of 97.47%. Additionally, we enhance model interpretability using Grad-CAM and LIME techniques, providing insights into the decision-making process and highlighting key features influencing classification. This framework offers a highly precise and efficient diagnostic tool, potentially aiding healthcare providers in early detection and diagnosis, particularly in resource-constrained environments.

Figures (11)

• Figure 1: Illustrative example of the dataset. Images a-g: Class 'MonkeyPox'. Images h–n: Class 'Others'.
• Figure 2: Illustrative example of the fourteen-fold augmentation on an image from the 'Monkeypox' class. Image a: original image. Images b–n: Augmented images.
• Figure 3: LightGBM leaf-wise tree growth Abbasniya2022.
• Figure 4: NGBoost is modular with respect to choice of base learner, distribution, and scoring rule Duan2020.
• Figure 5: Proposed methodological architecture.