CardiologyA company that uses ECG data to detect potential arrhythmias is developing a new way to predict atrial fibrillation.
Tuesday, European Heart Journal – Digital Health published a study Tested new algorithms from Cardiologs. The study, sponsored by the Paris-based company, found that the model accurately predicted the recent presence of AFib using only the first 24 hours recorded by a Holter monitor.
Founded in 2014, Cardiologs provides a device-agnostic software platform for arrhythmia detection. The platform is FDA-approved to screen for AFib and other cardiac arrhythmias to help physicians, rather than a standalone device.
The company develops its software using data uploaded by cardiologists from a variety of devices, including Holter monitors, smartwatches and ECG patches.In November last year, Philips Announce It is expanding its portfolio of cardiac solutions for hospital and outpatient settings with the acquisition of Cardiologs.
Dr. Jagmeet Singh, a cardiologist at Massachusetts General Hospital, led the company’s latest study. He and his team collected and de-identified Holter recordings from six independent diagnostic testing facilities in the US, EU, India, South Africa and the UK. They identified a set of training recordings, each lasting 7 to 15 days, in which no AFib was detected during the first 24 hours.
Using the first 24 hours of these recordings, the research team trained their algorithm to predict the presence or absence of AFib over the next 15 days. Using an external dataset not used during their development, they tested the algorithm and found that it could predict whether AFib would occur in the near future with an area under the receiver operating curve, sensitivity and specificity of 79.4 %, 76%, and 69%. The study also found that the model outperformed models that predicted recent AFib using a 12-lead ECG.
The algorithm gives “hope to high-risk patients who will benefit from aggressive treatment and AFib mitigation strategies,” Dr. Singh said in a statement. Press Releases.
Cardiologs developed the tool to help doctors identify more cases of AFib — a condition that often goes undetected and untreated, although CDC data It has been shown that this condition results in approximately 158,000 deaths each year. Patients typically require a 24-hour ECG to be diagnosed, but the diagnostic yield of this recording is low because of its short duration and the tendency to miss patients with infrequent episodes of AFib.
There are many other companies looking to use ECG data to improve AFib diagnosis, including apple and Rhythm.
In 2019, Apple released the results of a study showing that 84% of people who were notified of irregular heartbeats Apple Watch Found in AFib when notified.Now, Apple investigating Whether the ECG function of its watch can be used to detect other types of arrhythmias.
February, iRhythm release Kaiser Permanente researchers tested the findings of its Zio patch, which provides 3-14 days of uninterrupted cardiac monitoring. The findings showed that the Zio patch did a better job of detecting AFib than the 30-day event monitor.
Cardiologs’ research differs from research conducted by Apple and others because Cardiologs is the first company to use 24-hour Holter recordings to test an AI’s ability to predict AFib in the short term. A resting 12-lead ECG records the electrical activity of the heart from 12 electrical patches throughout the body over a short period of time and is a standard test for measuring the electrical function of the heart. These ECGs give doctors a more complete picture of heart activity than Holter monitors, but the latter provide longer-duration signals and therefore more input to AI models.
Photo: BrianAJackson, Getty Images
