Related Experiment Video
Updated: Aug 13, 2026
08:12
A Volumetric Method for Quantification of Cerebral Vasospasm in a Murine Model of Subarachnoid Hemorrhage
Published on: July 28, 2018
Magnetic resonance angiography: from anatomical knowledge modeling to vessel segmentation
N Passat1, C Ronse, J Baruthio
1Laboratoire des Sciences de l'Image, de l'Informatique et de la Télédétection (LSIIT), UMR 7005 CNRS-ULP, Bd S. Brant, BP 10413, F-67412 Illkirch Cedex, . passat@dpt-info.u-strasbg.fr
Medical Image Analysis
|January 3, 2006
Summary
This study introduces a novel approach for improved brain vessel segmentation using magnetic resonance angiography (MRA). By integrating a cerebral vascular atlas, the method enhances the accuracy of identifying vascular structures in MRA scans.
Area of Science:
- Medical Imaging
- Neuroscience
- Computer Vision
Background:
- Magnetic Resonance Angiography (MRA) is a non-invasive technique for visualizing cerebral vasculature.
- Current MRA vessel segmentation methods often yield unsatisfactory results, necessitating improved approaches.
- Accurate vessel segmentation is crucial for diagnosing pathologies, surgical planning, and functional brain analysis.
Purpose of the Study:
- To develop and evaluate a novel method for enhancing brain vessel segmentation from MRA data.
- To integrate high-level anatomical a priori knowledge into the segmentation process.
- To create a cerebral vascular atlas and utilize it for adaptive segmentation.
Main Methods:
- Developed a two-part method for Phase Contrast MRA (PC MRA) data.
- Part 1: Cerebral vascular atlas creation involving knowledge extraction (skeletonization-based vessel size determination), registration, and data fusion (topology-preserving non-rigid registration).
- Part 2: Segmentation using adaptive gray-level hit-or-miss operators guided by the atlas to tailor parameters (number, size, orientation) to specific vascular structures.
Main Results:
- An atlas was successfully created from an 18 MRA dataset.
- The atlas-guided segmentation method was applied to 30 MRA images.
- Results were compared against a traditional region-growing segmentation method, demonstrating potential improvements.
Conclusions:
- Integrating a cerebral vascular atlas with adaptive segmentation operators offers a promising solution for improving MRA-based brain vessel segmentation.
- The proposed method leverages anatomical knowledge to enhance segmentation accuracy.
- Further validation and refinement are warranted for clinical application.
