Authors: Ms Merlin Steffy M, Professor Dr. F. Ramesh Dhanaseelan, Associate Professor Dr. M. Jeya Sutha
Abstract: This research presents histotripsy is a non-invasive ultrasound technique used for precise tissue ablation, showing promise in treating conditions like kidney tumors. This project proposes a deep learning-based segmentation system using a Convolutional Neural Network (CNN) with a ResNet-18 backbone to identify ablated regions in ultrasound images automatically. The system is trained on phantom images and uses digital photographs as ground truth. In addition to image segmentation, the system overlays segmented zones, counts treatment pulses, and supports real-time monitoring significantly improving speed, accuracy, and clinical decision- making.
DOI: http://doi.org/10.5281/zenodo.15833637
