Liver Tumor Segmentation and Detection Using Deep Learning

Abstract

Liver cancer is one of the leading causes of cancer-related mortality worldwide, and early detection plays a pivotal role in improving patient survival rates. Accurate segmentation and detection of liver tumors from medical images such as CT and MRI scans are essential for diagnosis, treatment planning, and monitoring disease progression. This paper proposes an automated Liver Tumor Segmentation and Detection system using deep learning techniques, particularly Convolutional Neural Networks (CNN) and U-Net architecture. The system processes medical images through preprocessing steps including normalization, resizing, and noise reduction to enhance image quality. The proposed system achieves a Dice Similarity Coefficient of 0.91, sensitivity of 93.4%, and specificity of 95.7% on the LiTS benchmark dataset, significantly outperforming traditional image processing and machine learning baselines.