AI Tools Show Promise in Early Detection of Heart Dysfunction in Women of Reproductive Age

AI Tools Show Promise in Detecting Early Heart Dysfunction

Researchers at Mayo Clinic have made significant strides in the early detection of heart dysfunction in women of reproductive age using artificial intelligence (AI) tools. The study, published in the Annals of Family Medicine, evaluated the performance of an AI-enabled electrocardiogram (AI-ECG) and an AI-powered digital stethoscope in identifying early signs of heart problems 1

1

Study Design and Methodology

The cross-sectional pilot study, led by Dr. Anja Kinaszczuk and Dr. Demilade Adedinsewo, examined two groups of women aged 18 to 49, including those who were pregnant or within one year postpartum 2

2

1. Diagnostic cohort: 100 women scheduled for echocardiograms
2. Screening cohort: 100 women seen during routine primary care visits

All participants underwent two tests:

• A standard 10-second 12-lead electrocardiogram (ECG)
• A digital stethoscope recording capturing a 15-second, single-lead ECG and phonocardiogram

AI models analyzed these recordings to estimate the risk of left ventricular systolic dysfunction (LVSD).

Key Findings

The study revealed impressive results for both AI tools 3

3

• AI-ECG demonstrated an area under the curve (AUC) of 0.94
• AI digital stethoscope (Eko DUO) achieved an even higher AUC of 0.98

In the diagnostic cohort:

• 5% of women had LVSD
• AI-ECG showed a negative predictive value of 96.8%
• AI-stethoscope achieved a 100% negative predictive value

In the screening cohort:

• AI-ECG: 1% of women screened positive
• AI-stethoscope: 3.2% of women screened positive

Implications for Maternal Health

Dr. Adedinsewo emphasized the potential impact of these AI tools on maternal health: "Statistically, nearly half of pregnancies in this country are unplanned, and approximately 1% to 2% of women may have heart problems they don't know about. Our research findings suggest that these AI tools could be used to screen women before pregnancy, allowing for improved pregnancy planning and risk stratification, early treatment, and better health outcomes" 4

4

Related Stories

AI Revolutionizes Heart Disease Detection: From ECGs to Hidden Conditions

AI Model Identifies Female Patients at Higher Risk of Heart Disease Using ECG Analysis

AI Model Detects Multiple Heart Failure Types from Routine ECGs, Including Hard-to-Identify HFpEF

AI Software Significantly Improves Detection of Congenital Heart Defects in Prenatal Ultrasounds

Future Directions

This research builds upon earlier studies, including:

• A pilot prospective study evaluating AI digital tools to detect pregnancy-related cardiomyopathy among obstetric patients in the U.S.
• A pragmatic randomized clinical trial of women in Nigeria who were pregnant or had recently given birth

Further research is underway to explore the potential of using these technologies to screen for heart weakness in broader populations.

Technology Licensing

Mayo Clinic has licensed the underlying technology to:

• EKO Health for its digital stethoscope with embedded ECG electrodes
• Anumana for the 12-lead ECG

Mayo Clinic and some study authors have a financial interest in this technology, with any revenue received being used to support the clinic's not-for-profit mission in patient care, education, and research.