Machine Learning Based Diagnostic Paradigm in Viral and Non-Viral Hepatocellular Carcinoma

This study investigates the application of machine learning (ML) techniques for distinguishing viral and non-viral hepatocellular carcinoma (HCC), aiming to enhance diagnostic precision in clinical oncology. A comprehensive dataset comprising 204 patient records and 50 diverse features, including demographic, clinical, and laboratory parameters, was utilized to train and evaluate several classification algorithms, including Logistic Regression, Random Forest, Decision Tree, XGBoost, and AdaBoost. The study highlights differences in the performance of these algorithms, with ensemble-based methods demonstrating notable robustness in identifying complex patterns within the dataset. The findings suggest that machine learning approaches can significantly support clinicians in improving diagnostic accuracy, facilitating early detection, and informing personalized treatment strategies for HCC patients. This research underscores the potential of advanced ML methods to enhance decision-making processes in oncology by leveraging computational intelligence for more precise and efficient disease classification. Overall, the study provides a foundation for integrating machine learning into clinical workflows, emphasizing its role in advancing the diagnosis and management of both viral and non-viral forms of hepatocellular carcinoma.

Keywords: Machine Learning, Hepatocellular Carcinoma (HCC), Viral vs. Non-Viral Diagnosis, Classification Algorithms, Logistic Regression, Random Forest, Decision Tree, XGBoost, AdaBoost, Diagnostic Accuracy.