Decision Support System for Predicting Survivability of Hepatitis Patients

Area of Science:

• Medical informatics
• Machine learning in healthcare
• Virology

Background:

• Viral hepatitis poses a significant global health challenge, particularly in resource-limited regions, leading to severe complications like cirrhosis, liver failure, and cancer.
• Early diagnosis and treatment are crucial for reducing disease burden and transmission, making screening essential for meeting global health targets.
• Automated systems for hepatitis prediction are needed, but machine learning models have shown varied results with imbalanced datasets.

Purpose of the Study:

• To develop and evaluate a deep learning-based decision support system (DSS) for accurate binary classification of hepatitis patient survivability (mortality or survival).
• To utilize a bidirectional long/short-term memory (BiLSTM) model for predicting hepatitis using balanced datasets.

Main Methods:

• A deep learning approach employing a bidirectional long/short-term memory (BiLSTM) network was implemented.
• The model was trained and validated on balanced datasets for binary classification of hepatitis diagnosis.

Main Results:

• The proposed BiLSTM model demonstrated high performance metrics: 95.08% accuracy, 94% precision, 93% recall, and 93% F1-score.
• These results represent a significant improvement over existing methods in hepatitis detection and classification.

Conclusions:

• The bidirectional long/short-term memory (BiLSTM) model offers a superior approach for hepatitis classification compared to current methods.
• The developed DSS shows promise for improving the accuracy and efficiency of hepatitis diagnosis and patient management.