An AI-enhanced digital stethoscope has shown promising results in detecting peripartum cardiomyopathy, a potentially deadly form of heart failure that can occur late in pregnancy. A clinical trial conducted in Nigeria revealed that the AI-driven stethoscope was significantly more effective than traditional methods in identifying heart pump weakness linked to this condition.
The Screening for PEripartum Cardiomyopathies with Artificial Intelligence in Nigeria (SPEC-AI Nigeria) trial, a multicenter, open-label, randomized clinical trial, compared AI-guided ECG-based screening to clinical ECGs alone in identifying cardiomyopathy associated with left ventricular systolic dysfunction (LVSD) among pregnant and postpartum women. The study, published in Nature Medicine, involved nearly 1,200 pregnant women in Nigeria, a country with one of the highest rates of pregnancy-related heart failure.
Study Design and Methodology
Participants were randomized to either AI-guided screening or standard 12-lead clinical ECG in addition to usual care. The intervention arm received portable ECGs recorded with a digital stethoscope, 12-lead AI-ECG binary prediction for LVSD, and a confirmatory echocardiogram at baseline for validation of AI model performance. The AI algorithms used were based on convolutional neural networks trained on over 100,000 adults and were independent of all data collected in this study.
The primary outcome was the identification of cardiomyopathy, defined as LVEF <50% based on two-dimensional echocardiography. For the control arm, the number of participants with clinical recognition and documentation of LVSD (based on echocardiography) was used for the determination of the primary outcome, in keeping with current standard of care. The primary outcome in the intervention group was the number of participants with LVSD, as determined by a positive AI screen for LVSD confirmed by echocardiography at the time of ECG acquisition.
Key Findings
The AI-driven stethoscope was found to be 12 times more likely to spot heart pump weakness compared to traditional methods. Approximately 4% of women checked using the AI stethoscope were diagnosed with heart failure, compared to 2% of those provided traditional obstetric care.
According to lead researcher Dr. Demilade Adedinsewo, a cardiologist at Mayo Clinic, recognizing this type of heart failure early is crucial for the mother's health and well-being. Peripartum cardiomyopathy is often difficult to detect because its symptoms, such as shortness of breath and swelling in the feet or legs, mimic those of pregnancy.
Implications and Future Directions
Once detected, peripartum cardiomyopathy can be managed with medication. However, in severe cases, intensive care, mechanical heart pumps, or heart transplants may be necessary.
Researchers plan to further investigate the use of the AI stethoscope by Nigerian doctors and nurses, as well as its impact on pregnancy care in real-world scenarios. They also aim to conduct clinical trials in the United States, where peripartum cardiomyopathy affects approximately 1 in 2,000 women, and as many as 1 in 700 African-American women.
"This study provides evidence that we can better detect peripartum cardiomyopathy among women in Nigeria. However, there are more questions to be answered," Adedinsewo said.
