|
Briefing
|
- AI-Based Cardiovascular Risk Prediction – Google and Verily researchers found that assessing patients’ data extracted from retina, which determines their age, whether or not they smoke, and blood pressure, using machine learning can help predict risk of major cardiovascular disease
- Efficient, Accurate Results – Uses only retina scans to produce predictions with 70% accuracy, only slightly lower compared to traditional invasive methods, such as blood test, with 72%
- Algorithm Training – Fed machine learning algorithm with data from almost 300,000 patients to find underlying patterns by itself, then validated learnings on two datasets from more than 12,000 patients
- Potential Benefits – Can help develop new ways to analyze existing data, enabling medical professionals and patients to acquire more health-related information more efficiently
|
|
Accelerator
|
|
|
Sector
|
Healthcare/Health Sciences
|
|
Organization
|
Google Inc., Verily Life Sciences LLC
|
|
Source
|
-
Vincent, J., "Google’s new AI algorithm predicts heart disease by looking at your eyes",
-
Baig, E., "Google hopes AI can predict heart disease by looking at retinas",
-
Harwell, D. and Johnson, C., "In our eyes, Google’s software sees heart attack risk",
-
AcceleratingBiz analysis
|
|
Original Publication Date
|
February 19, 2018
|
