A new algorithm developed by researchers at Stanford can spot heart problems better than trained professionals.
Do you have a heart problem? Probably not, but especially for people over the age of 60 heart palpitations, fluctuations and the like can be signs of potential heart problems. But heart arrhythmias are tricky to spot: typically they can only be detected by a trained physician looking at an ECG, and that only happens if there are prior symptoms that make the patient go to the doctor in the first place.
In many cases, symptoms don't start appearing until it's almost too late, which means millions of Americans are walking around with undiagnosed heart conditions right now. These conditions are nearly impossible to detect, as they would require everyone to constantly walk around equipped with an ECG and trained physicians to review millions of hours of data.
One group of researchers is working to turn the impossible possible, by using machine learning. The researchers, from Stanford University, developed an algorithm that can spot heart irregularities at least as well as a physician, potentially revolutionizing the way we treat heart disease.
Their algorithm was trained on thousands of hours of ECG data and learned to accurately identify 14 different types of heart problems. In tests, the algorithm competed with six highly trained cardiologists and managed to outperform all of them.
The researchers plan to couple their software with a simple heart rate monitor, which can run continuously on people who are at high risk of heart disease. This system could give an early warning of potential problems long before noticeable symptoms start to appear, which could save thousands of lives.
Source: Stanford University via Engadget
Hi! I am a robot. I just upvoted you! I found similar content that readers might be interested in:
http://www.popularmechanics.com/science/health/a27251/cardiac-arrhythmia-machine-learning/
Downvoting a post can decrease pending rewards and make it less visible. Common reasons:
Submit