Study: AI-Enabled ECG Can Detect Aortic Stenosis Years Before Valve Replacement

AccurKardia is touting new study results indicating its artificial intelligence (AI)-enabled ECG technology, AK-AVS, can detect aortic stenosis years before patients require valve replacement and improve the prediction of clinical outcomes.

The study, “Validation and Longitudinal Trajectory Analysis of an AI-Based ECG Model for Aortic Stenosis: From Community Screening to Pre-TAVR Risk Stratification,” was led by principal investigator Dr. Matthew Segar, an electrophysiology Fellow at the Texas Heart Institute, and published in the European Heart Journal – Digital Health.

Aortic stenosis is one of the most common and serious valvular heart diseases among older adults and can lead to heart failure, hospitalization, and death if not diagnosed and treated early. The study evaluated AccurKardia’s AK-AVS AI model across both community-based populations and patients who underwent transcatheter aortic valve replacement (TAVR) at Baylor St. Luke’s Medical Center.

The study found that AK-AVS can detect aortic stenosis through routine ECGs up to four and a half years before TAVR intervention, potentially enabling earlier detection, easier-to-access monitoring, and better timing of intervention.

“This study demonstrates that AK-AVS could not only enable earlier detection of aortic stenosis, but it may also be a useful tool in surveillance and predicting outcomes,” said Dr. David Shavelle, chief of Cardiology for the MemorialCare Health System.

The findings also demonstrated that patients who screened positive for aortic stenosis using AI-ECG but did not yet show disease on echocardiography—traditionally labeled as “false positives”—still demonstrated a 4.4-fold increased risk of future aortic stenosis hospitalization during a median 6.2 years of follow-up. These results suggest the model may identify early electrical changes in the heart that occur before structural abnormalities are visible through conventional imaging.

Additionally, the study showed that AI-ECG trajectory patterns independently predict increased one-year mortality risk following TAVR, identifying patient risk that is not captured by widely used clinical risk scores such as the Society of Thoracic Surgeons (STS) and EuroSCORE models.

“By detecting subtle electrical remodeling patterns and tracking how they evolve over time, this technology has the potential to transform how clinicians screen, monitor, and risk-stratify patients, ultimately helping physicians intervene earlier and improve outcomes in aortic stenosis,” Dr. Segar stated.

Because ECGs are widely available, inexpensive, and routinely performed in clinical practice, AI-enhanced ECG analysis may expand access to screening and risk assessment across broad patient populations.

“Our next step is to complete real-world pilots of AK-AVS to demonstrate its ability to identify undiagnosed aortic stenosis patients and inform their subsequent treatment,” AccurKardia Chief Product Officer Moin Hussaini commented.

AccurKardia is an ECG-led diagnostics software company aiming to transform ECG data into a more powerful diagnostic tool and broad biomarker to improve patient outcomes. With initial applications in cardiology, the company offers cloud-based diagnostic tools, including AccurECG 2.0, a U.S. Food and Drug Administration (FDA)-cleared Class II software as a medical device (SaMD) for fully automated, near real-time ECG interpretation. AccurKardia is also advancing an investigational pipeline of ECG-based biomarkers, including AK+ Guard for hyperkalemia risk assessment using Lead I ECG and AK-AVS for aortic stenosis risk assessment, both of which have received FDA Breakthrough Device Designation.