RCE’s AI team trained a convolutional neural net on retrospective ECG data sets with ischemic vs non-ischemic classification. We then applied the held out model to prospective datasets from patients in cardiac observation unit. The accuracy of the model coupled with reliable data acquisition through a seamless ECG wearable is highly encouraging.
These initial efforts in artificial intelligence give us line of sight in the value of ischemia detection leveraging transfer learning. Check out our work here:
ECGDetect: Detecting Ischemia via Deep Learning
Coronary artery disease(CAD) is the most common type of heart disease and theleading cause of death worldwide[1]. A progressive state of this disease markedby plaque rupture and clot formation in the coronary arteries, also known as anacute coronary syndrome (ACS), is a condition of the heart ass…
arXiv.org