Google researchers presented recent findings in the online medical journal “Nature Biomedical Engineering” that help apply artificial intelligence to predict serious heart problems such as heart attacks or stroke in patients by examining images of the patient’s retina. The researchers say the method they have developed is as accurate in predicting heart disease as more invasive procedures. Google did however caution that the results from its research are preliminary and more work needs to be done.
The medical researchers were able to predict with a 70 percent accuracy whether a patient would have a heart attack or stroke in the next five years by examining a retinal image. Its research has shown a close correlation between retinal vessels and the risk of a major cardiovascular episode. The results are as accurate as other testing methods used that require blood being drawn to measure the patient’s cholesterol levels.
Google examined models based on data from 284,335 patients and validated two sets of data of 12,026 and 999 patients. One of the lead researchers on the project, Google’s Dr. Lily Peng said “the caveat to this is that it’s early (and) we trained this on a small data set.” She added that “we think that the accuracy of this prediction will go up a little bit more as we kind of get more comprehensive data. Discovering that we could do this is a good first step. But we need to validate.”
The research on predicting the likelihood of a heart attack or stroke was an offshoot of the original research done on work to predict eye disease, and whether a person smokes or has high blood pressure. Images showed that each cardiovascular risk factor has a distinct pattern, such as blood vessels for blood pressure and examination of the optic disc for other health predictions.
Harlan Krumhoz, a professor of medicine at Yale University Center for Outcomes Research and Evaluation, says that pattern recognition and use of images is one of the most successful areas for artificial intelligence, and that Google’s research is a good example of this success. “It will help us understand these processes and diagnoses in ways that we haven’t been able to do before,” he says. “And this is going to come from photographs and sensors and a whole range of devices that will help us essentially improve the physical examination and I think more precisely hone our understanding of disease and individuals and pair it with treatments.”
SOURCE: USA Today, “Google hopes AI can predict heart disease by looking at retinas,” by Edward C. Baig, February 19, 2018
MedTrust Transport provides emergent and non-emergent ambulance services in Charleston, Myrtle Beach and Georgetown, South Carolina. We have trained EMT personnel and a fleet of fully-equipped ambulances. We aim to provide compassionate and timely patient care.