Human monkeypox disease prediction using novel modified restricted Boltzmann machine-based equilibrium optimizer.

Area of Science:

• Medical Imaging
• Computational Biology
• Epidemiology

Background:

• The global health landscape faces challenges from emerging infectious diseases, including the recent rise in human monkeypox outbreaks.
• Accurate and timely diagnosis of monkeypox is crucial for pandemic prevention and control.
• Deep learning has shown promise in medical image analysis for disease prediction, including skin conditions.

Purpose of the Study:

• To develop and evaluate a novel deep learning methodology for image-oriented human monkeypox disease prediction.
• To assess the efficacy of a proposed model integrating Convolutional Block Attention Module (CBAM) and Modified Restricted Boltzmann Machine (MRBM) optimized by Equilibrium Optimizer (EO).

Main Methods:

• Utilized the Monkeypox Skin Lesion Dataset for training and validation.
• Applied image pre-processing techniques including resizing, normalization, and data augmentation.
• Employed CBAM for feature extraction and MRBM optimized by EO for prediction, focusing on error minimization.

Main Results:

• The proposed MRBM-EO model demonstrated superior performance in monkeypox prediction compared to existing models.
• Achieved significant improvements in Root Mean Square Error (RMSE), outperforming PSO-SVM, Xception-CBAM-Dense, ShuffleNet, and RBM.
• Showcased enhanced accuracy in disease prediction, surpassing benchmark models.

Conclusions:

• The developed deep learning model offers a promising tool for accurate and efficient human monkeypox prediction.
• The integration of CBAM and EO-optimized MRBM provides a robust framework for dermatological disease diagnosis.
• This approach can aid in early detection and management of monkeypox outbreaks, contributing to global health security.