Machine Learning + Health&care.
The most precious sound is that of a beating heart.
We have developed an ML-based system for automated segmentation of cardiac cycle stages and calculation of the probability of cardiac arrhythmia (atrial fibrillation).
We have taught the system to analyze cardiograms, calculate heartbeat data, and highlight deviations (discover disease patterns). Detecting the probability of arrhythmia helps discover irregularities that may be unnoticeable at a glance.
Such prompt analysis using Machine Learning helps to quickly diagnose diseases. It also draws the doctor’s attention to especially important cases by emphasizing areas that require extra care when prescribing a full check-up.
And, of course, it will be of use to regular users in the future. The patient uploads a cardiogram, the system analyzes it and, if deviations are detected, recommends a doctor’s appointment.
Would you like a house call from ML, M.D.?)
Read more about the project here: https://medrecord.io/atrial-fibrillation-detection-us..