Leveraging AI to prevent heart attacks and strokes offers a significant opportunity to transform healthcare and make it more productive, personalized, and accessible. Joining me today is Todd Villines, the Chief Medical Officer of Elucid, a pioneering medical technology company at the forefront of AI-powered heart attack and stroke prevention.
We discuss how Elucid's FDA-cleared product uses cutting-edge AI to analyze and characterize arterial plaque through CT scans and the innovative aspects of Elucid's algorithms. We explore the role of machine learning in Elucid's technology, from identifying risky plaques to using fractional flow reserve derived from CT without invasive procedures.
Tuning in, you’ll learn about the importance of high-quality data annotation and the rigorous validation process required to ensure accuracy across various scenarios and demographics. We also unpack the company's approach to annotating data, avoiding bias, and using diverse data sets. To discover how Elucid is making strides in cardiovascular health and paving the way for a healthier future, don’t miss this conversation with Todd Villines!
Key Points:
- Todd’s professional background and why he joined the team at Elucid.
- Elucid’s mission and how it leverages AI for medical technology.
- The role of machine learning in Elucid's technology.
- Developing a fractional flow reserve derived from CT analysis.
- Why data annotation is crucial for training the Elucid models.
- The importance of validation and how Elucid ensures the accuracy of its product.
- Challenges and limitations of working with CT images.
- How the company’s technology integrates into the existing clinical workflow.
- Metrics used to assess the impact of Elucid's technology.
- Intelligent design, diverse datasets, and avoiding bias in AI development.
- Discover Elucid's future outlook and its plans to expand.
Quotes:
“We’ve created proprietary algorithms that were trained by histology using traditional image processing techniques to recognize different types of plaque based on histology.” — Todd Villines
“In the field of medical imaging, using supervised machine learning and annotated data of very high quality and also generalizable to the clinical use case of your technology is vitally important.” — Todd Villines
“You can’t just go out and pick the very highest image quality to train your models or you’re going to end up with a very overfitted model that doesn’t generalize to the clinical use case.” — Todd Villines
“Just like any good clinical study, designing your AI technology is probably the most important thing. Spend the time upfront to get it right.” — Todd Villines
Links:
Resources for Computer Vision Teams:
LinkedIn – Connect with Heather.
Computer Vision Insights Newsletter – A biweekly newsletter to help bring the latest machine learning and computer vision research to applications in people and planetary health.
Computer Vision Strategy Session – Not sure how to advance your computer vision project? Get unstuck with a clear set of next steps. Schedule a 1 hour strategy session now to advance your project.
Computer Vision Advisory Services – Monthly advisory services to help you strategically plan your CV/ML capabilities, reduce the trial-and-error of model development, and get to market faster.
You Might Like
