Preloader Image animate3

Impact of heart disease and COVID-19 diagnosis, using artificial intelligence

  • 2021-01-03
  • Loreti

The U.S. Department of Health-funded researchers has found that artificial intelligence (AI) can help ECG interpretation arrhythmias diagnosis.

A powerful computer “studied” more than 90,000 ECG recordings, from which it “learned” to recognize patterns, form rules, and apply them accurately to future ECG readings. The computer became so “smart” that it could classify 16 different types of irregular heart rhythms, including atrial fibrillation (AFib). In fact, after just seven months of training, the computer-devised algorithm was as good—and in some cases even better than—cardiology experts at making the correct diagnostic call.

For example, in people with AFib, the heart’s upper chambers (the atria) contract rapidly and unpredictably, causing the ventricles (the main heart muscle) to contract irregularly rather than in a steady rhythm. This is an important arrhythmia to detect, even if it may only be present occasionally over many days of monitoring. That’s not always easy to do with current methods.

A total of 146 patients referred for evaluation of cardiac arrhythmia underwent simultaneous ambulatory ECG recording with a conventional 24-hour Holter monitor and a 14-day adhesive patch monitor. The primary outcome of the study was to compare the detection of arrhythmia events over total wear time for both devices. Arrhythmia events were defined as detection of any 1 of 6 arrhythmias, including supraventricular tachycardia, atrial fibrillation/flutter, pause greater than 3 seconds, atrioventricular block, ventricular tachycardia, or polymorphic ventricular tachycardia/ventricular fibrillation. McNemar's tests were used to compare the matched pairs of data from the Holter and the adhesive patch monitor.

Over the total wear time of both devices, the adhesive patch monitor detected 96 arrhythmia events compared with 61 arrhythmia events by the Holter monitor (P < .001).



The COVID-19 impact:

The pandemic outbreak of COVID-19 is estimated to add a massive boost to the growth of the cardiac monitoring market. The virus has impacted more than 200 economies and has affected around 14 million people globally. It attacks the respiratory system of the body leading to altered cardiac activity, thus it is important to monitor the vital signs and cardiac activity of patients with Covid-19 for any arrhythmia or other problems. Also, individuals with heart-related disorders are vulnerable to the infection leading to high concern for continuous monitoring activity of the heart.

To maintain the appropriate supply-demand curve various regulatory bodies have eased directives for the clearance of cardiac monitoring products.

 Due to the use of antimalarial drugs like hydroxychloroquine slight arrhythmia can be caused as a side effect. To avoid any further unfortunate issues such as cardiac arrest, close cardiac monitoring is necessary after the administration of hydroxychloroquine.


Main Benefits provided by the Holter monitors 2 or 3 ECG leads are an obvious advantage for both automatic algorithm analysis and physician interpretation. 

With ECG wearables growing in popularity, physicians now have more choices of ambulatory ECG monitoring solutions for their office-based patients. Assuming appropriate medical indications (subject to the patient’s insurance plan and coverage), insurer reimbursement of prescription-based ECG wearables is widely available and should not be a barrier to adoption. 


For Clinics

  • Time savings and removed bottlenecks with automatic data transfer and digital diary
  • More efficient use of resources by optimizing the recording time
  • The improved success rate of recordings

For Cardiologists

  • Increased quality control of recordings
  • Possibility to carry out long-term holtering with daily checks
  • Faster patient care

For Patients

  • Enables everyday life with comfortable, waterproof holter device
  • Optimization of recording time for each patient


Limitations remain, although new techniques show promise


AI still faces many practical challenges, though new techniques are emerging to address them. Machine learning can require large amounts of human effort to label the training data necessary for supervised learning. 

Obtaining data sets that are sufficiently large and comprehensive to be used for training—for example, creating or obtaining sufficient clinical-trial data to predict healthcare treatment outcomes more accurately—is also often challenging.

  • 1