Can machine learning improve the diagnosis and treatment of heart disease? It’s too soon to know for certain, but many promising tools are in the works


Assuming you have a cell phone, the photos you take with it are naturally improved by programming that has been "instructed" to cause your image to seem more honed and more brilliant by dissecting a great many comparative photographs. This kind of AI — a critical part of man-made consciousness — is like what's going on in cardio­vascular care (see "Figuring out computerized reasoning"). However, rather than examining photographs, the projects break down pictures from chest CT filters, ultrasound pictures of the heart (echocardiograms), and drawings of the heart's electrical action (electrocardiograms).


 

"We currently have apparatuses that can take in immense measures of many sorts of information, including numbers, pictures, and even sounds," says Dr. Michael Lu, head of man-made consciousness for the Cardio­vascular Imaging Research Center at Harvard-subsidiary Massachusetts General Hospital. AI models made with this information can distinguish designs that might be useful to doctors arrive at additional exact conclusions about diagnosing and treating coronary illness.