We were interested to read the paper by Attia et al.1 which demonstrated the value of electrocardiogram enabled stethoscopes (ECG-Scope). Their findings show potential in the utilization of artificial intelligence (AI) algorithms in conjunction with a single lead ECG-Scope to identify left ventricular dysfunction (LVSD). A clinical pathway such as this may speed up diagnosis and potentially improve patient outcomes.
Bachtiger et al.2 utilized ECG-Scope differently to Attia et al., choosing to auscultate at standard valvular landmarks rather than modified ECG electrode positions. Bachtiger et al. achieved best results over the pulmonary position, while Attia et al. indicated a V2 position when supine was best. Interestingly, they both achieved consistently strong results with their handheld variants, and this involved patients touching electrodes on the ECG-Scope with their fingers/thumbs. For both, handheld was shown to be second best, and for Attia et al., the area under curve (AUC) remained above 0.84 in detecting reduced ejection fraction (EF), irrespective of patient position.
One of the limitations of ECG-Scope going forward is the cost of widespread implementation of new technology to make an impact in non-specialist services. With the demand for smart wearables continuing to escalate, the results from Kwon et al.3 provide an interesting alternative. Making use of already available smartwatches from Apple and Samsung, they were able to generate a 12-lead ECG from a smartwatch ECG (Lead I and II). This AI model was able to demonstrate potential for use as a screening tool (AUC 0.925–0.946).
Han et al.4 also made use of multiple lead-based AI enabled ECG algorithms on smartwatch ECG data. Data showed that model performance improved as the number of leads increased, with 12-lead being the best (AUC 0.880). Making use of consumer wearables can reduce the burden of technological upgrade costs needed to make this model of diagnosis work. It also indicates that all AI algorithms should be made to generate 12-lead ECGs, including with ECG-Scopes.
In conclusion, AI algorithms on ECGs obtained from non-conventional sources have shown potential in being able to diagnose earlier and potentially improve patient outcomes in LVSD. Utilizing consumer wearables in conjunction with existing stethoscopes may help to realize real-world benefits sooner, especially in a primary care setting, compared with investment in new clinician tools. Work should be done to ensure algorithms aim for 12 leads to improve model performance.
Contributor Information
Nishil Patel, North Middlesex University Hospital, Sterling Way, London N18 1QX, United Kingdom.
Salaheldin Agamy, Royal Free Hospital, London, United Kingdom.
Mahmood Ahmad, Royal Free Hospital, London, United Kingdom.
Data availability
No new data has been utilised in this letter. Results are taken from references and indicated with citation.
References
- 1. Attia Z, Dugan J, Rideout A, Maidens JN, Venkatraman S, Guo L, Noseworthy PA, Pellikka PA, Pham SL, Kapa S, Friedman PA, Lopez-Jimenez F. Automated detection of low ejection fraction from a one-lead electrocardiogram: application of an AI algorithm to an electrocardiogram-enabled Digital Stethoscope. Eur Hear J 2022;3:373–379. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Bachtiger P, Petri CF, Scott FE, Ri Park S, Kelshiker MA, Sahemey HK, Dumea B, Alquero R, Padam PS, Hatrick IR, Ali A, Ribeiro M, Cheung W-S, Bual N, Rana B, Shun-Shin M, Kramer DB, Fragoyannis A, Keene D, Plymen CM, Peters NS. Point-of-care screening for heart failure with reduced ejection fraction using artificial intelligence during ECG-enabled stethoscope examination in London, UK: a prospective, observational, multicentre study. Lancet Digit Heal 2022;4:e117–e125. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Kwon JM, Jo YY, Lee SY, Kang S, Lim SY, Lee MS, Kim K-H. Artificial intelligence-enhanced smartwatch ECG for heart failure-reduced ejection fraction detection by generating 12-lead ECG. Diagnostics (Basel) 2022;12:654. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Han C, Song Y, Lim HS, Tae Y, Jang JH, Lee BT, Lee Y, Bae W, Yoon D. Automated detection of acute myocardial infarction using asynchronous electrocardiogram signals-preview of implementing artificial intelligence with multichannel electrocardiographs obtained from smartwatches: retrospective study. J Med Internet Res 2021;23:e31129. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
No new data has been utilised in this letter. Results are taken from references and indicated with citation.