TY - JOUR AU - kapoor, Ishani AU - Mittal, Deepti PY - 2016/12/30 Y2 - 2024/03/28 TI - An Efficient Method for Hepatic Cyst Segmentation on Ultrasound Images JF - British Journal of Healthcare and Medical Research JA - BJHMR VL - 3 IS - 6 SE - Original Articles DO - 10.14738/jbemi.36.2550 UR - https://journals.scholarpublishing.org/index.php/JBEMi/article/view/2550 SP - 55 AB - <p>Hepatic cysts are usually rare complications but when they occur they may be fatal. Thus the radiologist should be able to monitor the cyst regularly and for doing so they need some parameters to monitor. The size of the cyst being one of the parameters that can be monitored after the segmentation of the cysts. The present work is focused on the designing of a modified method that can segment hepatic cyst on ultrasound images. This proposed method is designed by modification in geometric method proposed by Chan and Vese [1] which is combination of active contour and level set method. The work is performed on seventeen clinically acquired ultrasound images, which comprises of 6 small, 5 large, 4 multiple and 2 atypical type cyst images. The modified method is validated by both qualitative and quantitative criteria. The performance of proposed method can be expressed in terms of accuracy, sensitivity and specificity showing 97.9%, 77.3% and 90.7% for small cysts, 94.1%, 68.0% and 80.0% for large cysts, 98.2%, 96.3% and 98.5%for multiple cysts and for 95.7%,  68.9% and 91.3% atypical cysts respectively. The results demonstrate that the modified method can segment effectively hepatic cyst on ultrasound images and therefore can be used by the radiologist for diagnostic purposes.</p> ER -