Interactive Medical Image Segmentation using Active Contour with Improved F Energy in Level-Set Tuning
Authors
Abstract
Segmentation is a fundamental element in medical image processing (MIP) and has been extensively researched and developed to aid in clinical interpretation and utilization. This article discusses a method for segmenting abnormal masses or tumors in medical images that is both robust and effective. We suggested a method based on Active Contour (AC) and modified Level-set techniques to detect malignancies in magnetic resonance imaging (MRI), mammography, and computed tomography (CT). To segment malignant masses, the active contour approach, the energy function, the level-set method, and the proposed F function are employed. The system was evaluated using 160 medical images from two databases, including 80 mammograms and 80 MRI brain scans. The algorithm for segmenting suspicious segments has an accuracy, recall, and precision of 96.25%, 95.60%, and 95.71%, respectively. By adding this technique into tissue imaging devices, the accuracy of diagnosing images with a relatively large volume that are evaluated fast is increased. Cost savings, time savings, and high precision are all advantages of the approach that set it apart from similar systems.
