Google researchers said this approach was 70 percent accurate in identifying patients who would suffer a heart attack or other major cardiovascular problem within five years and those who would not, USA Today reported.
That rate is similar to blood tests to measure cholesterol levels.
For the study, the researchers used models that were based on data from 284,335 patients and validated using two separate data sets of 12,026 and 999 patients. The findings were published Monday in the online journal Nature Biomedical Engineering.
The study authors noted that much more research needs to be done.
"The caveat to this is that it's early, (and) we trained this on a small data set," lead researcher Lily Peng told USA Today. "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."