Remote monitoring devices like insertable cardiac monitors have already massively cut down on the amount of time that clinicians must spend actively tracking patients’ conditions—but French startup Implicity is hoping to make those devices even smarter.
The company, which also has a headquarters in Cambridge, Massachusetts, has been granted 510(k) clearance by the FDA for its artificial intelligence software. The AI algorithm is designed to analyze ECG data gathered by those ICMs, also known as implantable loop recorders, or ILRs.
The ILR ECG Analyzer algorithm sifts through the collected heart rhythm data to spot instances of atrial fibrillation and other arrhythmias with what Implicity says is greater accuracy than the implanted monitors themselves.
RELATED: Abbott debuts smart monitor that fishes for abnormal heart rhythms in the data stream
ICMs are notorious for generating false-positive arrhythmia alerts, each of which must be further reviewed and confirmed (or rejected) by clinicians. A 2020 study estimated that between 46% and 86% of instances flagged by Medtronic’s Reveal Linq device, for one, are misdiagnoses.
According to a study of Implicity’s technology published in October, however, applying the machine learning algorithm to ECG data collected by Medtronic’s ICMs decreased the false positive rate by nearly 80%. The AI looked at more than 2,800 arrhythmia alerts from 370 patients’ devices and ultimately reclassified 43% of them as normal rhythm.
Additionally, the algorithm maintained 99% sensitivity in spotting true-positive cases of arrhythmia, the study found.
“The true promise of remote monitoring is that we can see patients who need to be seen earlier. Even if the patient is not experiencing a symptom, we can detect their condition and let them know they need to be seen,” said Niraj Varma, M.D., Ph.D., one of the study’s authors and a consultant electrophysiologist at the Cleveland Clinic. “Implicity’s innovative solutions will enable us to direct our attention to clinically actionable data better so we can do just that.”
The algorithm was cleared for use alongside all but the latest model in Medtronic’s widely used line of ICMs, including the Reveal Linq, Reveal XT and Reveal DX devices.
RELATED: Medtronic nets heartening FDA nod for AI algorithms to detect afib, arrhythmia
Implicity’s technology augments Medtronic’s own, as the medtech giant recently introduced FDA-cleared AI of its own to improve arrhythmia detection by only its Linq II monitor, the newest addition to its ICM portfolio.
The pair of AccuRhythm algorithms scored FDA clearance over the summer. Like Implicity’s version, Medtronic’s cloud-based AI reads cardiac data collected by the Linq II implanted devices, then automatically alerts a patient’s cardiologist and care team when it spots signs of afib or asystole.
A study of the afib-specific AccuRhythm software found that it was able to reduce false alarms by about 74%, while also demonstrating a sensitivity of more than 99%. The algorithm designed to spot asystole, meanwhile, slashed the number of false-positive alerts by more than 97%.