OHIF / Viewers

OHIF zero-footprint DICOM viewer and oncology specific Lesion Tracker, plus shared extension packages
https://docs.ohif.org/
MIT License
3.35k stars 3.37k forks source link

[3.10-GC] Implement High-Performance WebGPU-Based 3D Grow Cut Segmentation Toolkit #4512

Open sedghi opened 3 days ago

sedghi commented 3 days ago

Purpose
This feature introduces a high-performance, WebGPU-accelerated 3D Grow Cut segmentation tool, designed for fast and efficient segmentation of large image volumes. By leveraging WebGPU, this tool provides rapid processing and supports both manual and automatic segmentation modes, giving users flexibility to control segmentation boundaries or opt for a streamlined, one-click segmentation experience.

Why This Matters
3D segmentation is a computationally intensive process, especially with high-resolution medical imaging data. Using WebGPU for parallel processing dramatically increases segmentation speed, making advanced analysis more accessible and less time-consuming. This tool is particularly beneficial in contexts like PET imaging, where rapid, precise segmentation is critical for identifying regions of interest.

Key Changes

Impact on Users and Developers