A novel artificial intelligence (AI) model that analyzes electrocardiogram (ECG) signals from smartwatches may help detect left ventricular systolic dysfunction (LVSD), a condition where the heart's main pumping chamber does not contract effectively, according to preliminary research presented at the American Heart Association's Scientific Sessions 2026. The findings suggest a potential role for consumer wearables in flagging people who need further heart screening, but also reveal important limitations that must be addressed before such technology can be relied upon clinically.
LVSD, defined in the study as a left ventricular ejection fraction of 40% or less, can be present without symptoms and, if untreated, may lead to symptomatic heart failure. The study, led by Dr. Hak Seung Lee of Medical AI Co., Ltd. and the Digital Healthcare Institute in Incheon, South Korea, explored whether AI analysis of smartwatch ECGs could identify individuals who might benefit from an echocardiogram. As Dr. Lee noted, "Some people have reduced heart pumping function without noticeable symptoms," underscoring the need for accessible screening tools. The abstract was presented as late-breaking science at the meeting.
In the first part of the study, 2,211 adults scheduled for an echocardiogram had ECGs recorded on both an Apple Watch and a Samsung Galaxy Watch during the same clinic visit. The AI model distinguished people with and without LVSD about equally well on both devices. However, when a predetermined AI-score cutoff was applied, 200 participants (9%) were classified as positive on one watch and negative on the other, indicating that brand-specific differences could affect reliability. This inconsistency is a critical consideration for real-world use, as consumers may use different devices interchangeably.
A second group of 146 adults recorded ECGs at home, with 108 participants providing at least five days of recordings in the first week. When the AI required positive results on at least 2 of 3 consecutive days, false-positive classifications were halved among those without LVSD (from 18 to 9 among 70 participants), but two additional cases of LVSD were missed compared with a single-positive-day rule. Among all 146 home-recording participants, 81% of those with consistently positive results had LVSD confirmed by echocardiography, versus about 6% of those with consistently negative results. However, 49% had mixed results, and 40% of that group had confirmed LVSD, highlighting the complexity of interpreting repeated measurements. "A single smartwatch ECG is a snapshot, and repeated recordings can give different results," Dr. Lee said. "Requiring repeated positive results reduced false positives in our study, yet it also missed some additional cases that should have been positive."
The study has several limitations, including its single-center design in South Korea and a participant pool referred for echocardiography, which may not reflect the general population. The repeat-testing analysis included only 108 participants and is considered exploratory. The researchers emphasize that larger studies are needed to determine how best to use these results to guide further evaluation. Moreover, the study did not assess whether AI-guided screening leads to earlier treatment or improved outcomes. For now, smartwatch ECG readings cannot replace traditional screening exams.
The findings were presented at the American Heart Association's Scientific Sessions 2026, a premier global exchange of cardiovascular science. The association provides resources on ejection fraction heart failure measurement and initiatives like Get With The Guidelines® - Heart Failure to support clinical practice. As the technology evolves, the potential for AI-enhanced smartwatch ECGs to enable earlier detection of LVSD remains promising, but the variability across brands and recording strategies must be resolved through further research. The full release is available via www.newmediawire.com.
