Optimizing HCV Disease Prediction in Egypt: The hyOPTGB Framework.

Area of Science:

• Medical Informatics
• Machine Learning
• Epidemiology

Background:

• Egypt faces a high prevalence of Hepatitis C Virus (HCV) infection, driven by factors like injection drug use and inadequate healthcare practices.
• Accurate prediction of HCV is crucial for effective public health interventions and resource allocation in high-prevalence regions.

Purpose of the Study:

• To develop and evaluate a highly accurate machine learning model for predicting HCV infection in Egypt.
• To compare the performance of the proposed hyOPTGB model against other established machine learning algorithms using a relevant dataset.

Main Methods:

• A novel hyOPTGB model was developed, utilizing an optimized gradient boosting classifier with hyperparameter tuning via the OPTUNA framework.
• Data preprocessing involved Min-Max normalization and feature selection using the forward selection (FS) wrapped method.
• The model was trained and evaluated on a dataset of 1385 instances and 29 features from the UCI machine learning repository.

Main Results:

• The hyOPTGB model achieved a superior accuracy of 95.3%, outperforming other models including Decision Tree (DT), Support Vector Machine (SVM), Dummy Classifier (DC), Ridge Classifier (RC), and Bagging Classifier (BC).
• Performance was further validated by comparing against existing models applied to the same dataset, demonstrating consistent efficacy.
• Key performance metrics such as accuracy, recall, precision, and F1-score were used to assess the system's effectiveness.

Conclusions:

• The hyOPTGB model demonstrates significant potential as an effective tool for predicting HCV infection in Egypt.
• The optimized gradient boosting approach offers a promising direction for improving diagnostic accuracy in public health.
• Accurate HCV prediction can support targeted interventions and reduce the disease burden in high-prevalence populations.