Vascular Territory Segmentation Software
Patent Number: US9196037B2
Executive Summary:
General Description: Methods, systems, computer programs, circuits and workstations are configured to automatically segment an image volume (MRI, etc.) into separate vascular territories using mutual clustering in image and label space to generate at least one two-dimensional weighted cerebral blood flow (CBF) territory map of color-coded source artery locations using an automated vascular segmentation process to identify source locations using mutual connectivity in both image and label space. For example, right internal carotid artery may be depicted in blue, left internal carotid artery in red, and right vertebral artery in cyan, and left vertebral artery in yellow and amount of perfusion can be indicated typically in brightness and/or opacity. This automated vascular territory mapping is used for resolving source locations and to determine if multiple sources in different perfusion territories are from a single artery. This may be particularly useful for MRI brain scans for evaluation of large artery diseases, cerebral vascular disease, and carotid stenosis and may also be useful for stroke, especially thromboembolic stroke, and/or for evaluation of treatments or clinical trials.
Future Directions:
Strengths:
Weaknesses:
Patent Status:
Inventor Bio: Youngkyoo Jung
http://www.wakehealth.edu/Faculty/Jung-Youngkyoo.htm
Executive Summary:
- Invention Type: Prognostic and Diagnostic
- Patent Status: Grant
- Patent Link: https://patents.google.com/patent/US9196037B2/
- Research Institute: Wake Forest University Health Sciences
- Disease Focus: Vascular Disease
- Basis of Invention: Biomedical image inspection that automatically segments an image volume into separate vascular territories using mutual clustering in image and label space to create a clinically acceptable method for visualization and quantification of perfusion territories from major feeding arteries in the brain.
- How it works: Automated vascular territory segmentation method to identify arterial source location applying a three-dimensional connectivity clustering analysis to a brain image to electronically map source locations, electronically determining if multiple territories are associated with a single source, and mapping cerebral blood flow
- Lead Challenge Inventor: Youngkyoo Jung
- Development Stage: Preclinical
- Novelty:
- Vascular territory mapping method requiring minimal operator intervention and simple post-processing routines that are suitable for clinical implementation
- Voxel-wise quantitative mapping of blood flow in the brain
- Clinical Applications:
- Mapping of cerebral perfusion territories
- Evaluating large artery diseases, cerebral vascular disease, and carotid stenosis and may also be useful for stroke, especially thromboembolic stroke, and/or for evaluation of treatments or clinical trials.
General Description: Methods, systems, computer programs, circuits and workstations are configured to automatically segment an image volume (MRI, etc.) into separate vascular territories using mutual clustering in image and label space to generate at least one two-dimensional weighted cerebral blood flow (CBF) territory map of color-coded source artery locations using an automated vascular segmentation process to identify source locations using mutual connectivity in both image and label space. For example, right internal carotid artery may be depicted in blue, left internal carotid artery in red, and right vertebral artery in cyan, and left vertebral artery in yellow and amount of perfusion can be indicated typically in brightness and/or opacity. This automated vascular territory mapping is used for resolving source locations and to determine if multiple sources in different perfusion territories are from a single artery. This may be particularly useful for MRI brain scans for evaluation of large artery diseases, cerebral vascular disease, and carotid stenosis and may also be useful for stroke, especially thromboembolic stroke, and/or for evaluation of treatments or clinical trials.
Future Directions:
- Reproducing results
Strengths:
- Novel approach to addressing cerebral blood flow and source identification
- Automated approach improves accessibility and readability for point of care decision making for clinicians
Weaknesses:
- Requires trials and validation of results
Patent Status:
- Priority date: 2013-03-15
- Filing date: 2014-09-18
- Publication date: 2015-11-24
- Grant date: 2015-11-24
Inventor Bio: Youngkyoo Jung
http://www.wakehealth.edu/Faculty/Jung-Youngkyoo.htm
