About this document
3D CNN for Aneurysm Segmentation by IzzHyuk is a document available to read on EtoBox.
The document discusses a 3D convolutional neural network developed for abdominal aortic aneurysm segmentation from computed tomography angiography scans. It aims to standardize aneurysm diameter measurements and allow for more complex analysis of aneurysm evolution. The network is validated on preoperative and postoperative data and achieves a mean diameter difference of 3.3 mm and Dice similarity coefficient of 87%.
- Author
- IzzHyuk
- Language
- EN