Computer-Aided Diagnosis for Liver Tumor Feature Extraction

Liver tumors are a significant health concern, and their timely and accurate diagnosis is crucial for effective treatment planning. Computer-aided diagnosis (CAD) systems have shown promise in assisting medical professionals with the detection and characterization of liver tumors from medical imaging data. The objective was to classify the input CT images as normal or abnormal, based on the presence of liver tumors. For this purpose, three classifiers, Gradient Boosting Machines (GBM), K-nearest Neighbor (KNN), and Support Vector Machine (SVM), were utilized. The findings demonstrate the effectiveness of the proposed segmentation and classification approach in distinguishing between normal and abnormal CT images and indicate that KNN is the most accurate classifier among the three models evaluated. This research contributes to the advancement of computer-aided diagnosis for liver tumor detection and could have implications for improved medical decision-making and patient care.