Deep Learning Framework Developed for Liver Segmentation in 3D CT Images

This announcement presents a new deep learning-based framework designed to improve liver segmentation from 3D CT volumes, addressing challenges such as anatomical variability and indistinct organ boundaries. The approach uses a 3D probabilistic shape map derived from a reference atlas dataset, aligned with input CT scans through 3D affine registration to enhance spatial correspondence and segmentation accuracy.

Originally published by gnews