Early Hepatitis Detection using Convolutional Neural Network and GeneticAlgorithm

Abstract

The early detection of hepatitis is important for effective treatment and the prevention of severe liver damage. Existing diagnostic methods often struggle in identifying early-stage hepatitis due to challenges such as data complexity, imbalanced datasets, and inefficient feature extraction methods in traditional models. This research introduces a new approach for early hepatitis detection using a Convolutional Neural Network (CNN) optimized with a GeneticAlgorithm (GA) in order to gain better performance. Leveraging the hepatitis dataset, the study addresses the critical challenge of early diagnosis, which is key to improving patient outcomes and managing disease progression. The methodology trains a CNN model on patient data, with the GAemployed to fine-tune hyperparameters for optimal performance. The model achieved a high accuracy of more than 97% in correctly identifying early-stage hepatitis. High AUC-ROC scores further validate the model's reliability and effectiveness. Compared to other machine learning and deep learning models, the GA-optimized CNN consistently outperformed its counterparts, highlighting its potential as a valuable tool in clinical settings. This research emphasizes the significant role and impact advanced AI techniques can play in medical diagnostics, particularly in the early detection of diseases like hepatitis, where timely intervention is critical.