Related Experiment Video
Updated: May 30, 2025
10:09
Optimization of a Quantitative Micro-neutralization Assay
Published on: December 14, 2016
21.0K
An optimized ensemble grey wolf-based pipeline for monkeypox diagnosis
Ahmed I Saleh1, Asmaa H Rabie1, Shimaa E ElSayyad1,2
1Computers and Control Systems Engineering Department, Faculty of Engineering, Mansoura University, Mansoura, 35516, Egypt.
Scientific Reports
|January 30, 2025
Summary
A new hybrid AI model offers rapid and accurate automatic monkeypox diagnosis. This advanced system achieves high accuracy, demonstrating its potential for early detection of infectious diseases like monkeypox.
Area of Science:
- Medical Diagnostics
- Artificial Intelligence
- Infectious Disease Surveillance
Background:
- The emergence of monkeypox virus post-coronavirus pandemic necessitates advanced diagnostic tools.
- Current diagnostic methods may lack the speed and efficiency required for emerging infectious diseases.
Purpose of the Study:
- To develop a hybrid AI architecture for automated monkeypox diagnosis.
- To enhance diagnostic speed and accuracy using optimized feature selection and ensemble classification.
Main Methods:
- Utilized a modified grey wolf optimization for feature selection and weighting.
- Employed an ensemble of classifiers with a confusion-based voting scheme.
- Evaluated performance on public datasets with varying training sample sizes.
Main Results:
- Achieved 98.91% accuracy with a 5.5-second testing runtime.
- Demonstrated superior performance compared to existing literature approaches across multiple metrics.
- Validated generalizability on external monkeypox and COVID-19 datasets with 99.00% and 98.00% accuracy, respectively.
Conclusions:
- The proposed automatic monkeypox diagnostic system (AMDS) shows high accuracy and efficiency.
- The hybrid AI approach offers a robust solution for diagnosing emerging infectious diseases.
- The model's generalizability across different viral diseases is confirmed.
