Machine Learning Approaches for the Prediction of Hepatitis B and C Seropositivity

Area of Science:

• Hepatology
• Medical Informatics
• Public Health

Background:

• Identifying patients at risk for Hepatitis B (HBV) and Hepatitis C (HCV) infections presents a significant clinical and public health challenge.
• Early detection and risk stratification are crucial for managing viral hepatitis.
• Machine learning (ML) offers potential solutions for improving diagnostic accuracy and patient management.

Purpose of the Study:

• To evaluate and compare the predictive performance of four machine learning models (SVM, RF, NB, KNN) for HBV and HCV status.
• To assess the utility of ML in identifying patients at risk for viral hepatitis.
• To compare the efficacy of different ML algorithms in predicting HCV versus HBV infection.

Main Methods:

• A prospective cohort screening study was conducted in Romania from January to November 2022.
• Adult patients undergoing viral hepatitis screening provided clinical data via a structured survey.
• Four ML models—Support Vector Machine (SVM), Random Forest (RF), Naïve Bayes (NB), and K-Nearest Neighbors (KNN)—were trained and evaluated on the collected data.

Main Results:

• All ML models demonstrated superior performance in predicting HCV status compared to HBV status.
• The K-Nearest Neighbors (KNN) algorithm achieved the highest accuracy for HCV prediction (98.1%).
• Support Vector Machine (SVM) and Random Forest (RF) models showed high accuracy for HCV (97.6%), while Naïve Bayes (NB) achieved 95.7%. Predictive accuracies for HBV ranged from 78.2% to 97.6%.

Conclusions:

• Machine learning models show significant promise as tools for predicting HCV infection.
• These ML models can aid in the risk stratification of adult patients within viral hepatitis screening programs.
• Further research may refine ML applications for both HBV and HCV risk assessment and early detection.