The early and accurate diagnosis of Hepatitis C Virus (HCV) is crucial for timely medical intervention and improved patient outcomes. This study evaluates the effectiveness of various machine learning (ML) algorithms in classifying patients into different diagnostic categories, including healthy blood donors and stages of Hepatitis C such as Fibrosis and Cirrhosis. The models analyzed include Logistic Regression, Naive Bayes, Decision Tree, Random Forest, Gradient Boosting, XGBoost, Support Vector Classifier (SVC), and LightGBM. The dataset underwent preprocessing and was balanced using the SMOTE technique to ensure fair evaluation. Performance was assessed using key metrics such as accuracy, F1-score, and AUC-ROC. The results indicate that SVC, LightGBM, and Random Forest outperformed other models, demonstrating high accuracy and balanced precision and recall. Logistic Regression also showed solid performance, making it a reliable, interpretable option. In contrast, simpler models like Decision Tree struggled to maintain competitive scores across all metrics. The use of SMOTE to balance the dataset proved effective in addressing class imbalance, enhancing the models’ overall performance. Additionally, feature importance analysis provided insights into the most significant variables influencing the classification process, aiding in the understanding of key diagnostic indicators. The findings emphasize the robustness and applicability of advanced ensemble methods and support vector machines for medical diagnostics, ensuring more accurate classification and supporting healthcare providers in early detection and management.