helping patients with coronary heart disease
Coronary heart disease is the single largest cause of death in the United States. Currently, examining coronary angiograms largely relies on visual examination by physicians, making interpretation prone to human error. As a result, some patients are exposed to unnecessary, invasive procedures while other patients fail to receive appropriate treatments.
University of Michigan faculty Kayvan Najarian, Ph.D., and Brahmajee Nallamothu, M.D., are developing a system to utilize advanced image processing and machine learning techniques to analyze coronary angiograms independent of human input.
Preliminary results show that the new platform has the potential to zero in on a specific part of the vascular tree for more intense study, estimate the width of the blood vessels, and quantitatively identify the presence and percentage of blockages in each blood vessel. This objective data will help cardiologists with clinical decision making regarding the need for a stent or other interventional treatments.
The fully-automated computer platform could lead to new quality review and decision-support tools that are scalable and cost-effective, while at the same time promoting novel educational and support resources that will result in increased surgical accuracy.
Significant Need
Existing computer-based techniques that assist cardiologists in interpreting coronary angiograms require significant human input, create time delays that disrupt workflow, and consume substantial resources.
Compelling Science
A fully-automated computer platform that utilizes advanced image processing and machine learning techniques to analyze coronary angiograms.
Competitive Advantage
Currently, angiogram examination largely relies on visual examination by human operators. A fully-automated system guides real-time clinical diagnosis as well as quality assurance without the limitations of human input.
Overall Commercialization
- Intellectual Property: Invention disclosure filed.
- Commercialization Strategy: Form a start-up company.
- Regulatory Pathway: Follow a 510K pathway.
- Engage Investors: There is interest from large venture capital group.
- Product Launch Strategy: License to an existing image analysis or medical auditing group.
Milestones
- Collect and annotate coronary angiograms
- Further develop and optimize the “Stabilization and Motion Artifact Removal” algorithm
- Further improve the “Detection of Artifacts” algorithm, develop and improve the “Segmentation of Blood Vessels” algorithm, and further improve and optimize the “Detection of Vessel Thickness/Stenosis algorithm
- Validate results
- Design graphical user interface
- Work with partner company to address any potential issue towards commercialization and regulatory procedures