Authors: Demilade A Adedinsewo, Andrea Carolina Morales-Lara Morales Lara, Patrick W Johnson, Mikolaj A Wieczorek, Mayo Clinic, Jacksonville, FL; Jennifer Dugan, Xiaoxi Yao, Zachi I Attia, Mayo Clinic, Rochester, MN; Francisco Lopez-Jimenez, Mayo Clinic Coll Med, Rochester, MN; Paul A Friedman, Peter A Noseworthy, Mayo Clinic, Rochester, MN; Rickey E Carter, Mayo Clinic, Jacksonville, FL; SPEC-AI Nigeria Investigators
Abstract: Nigeria has the highest reported incidence of peripartum cardiomyopathy worldwide. A delay in the diagnosis and treatment of cardiomyopathy significantly contributes to excess maternal deaths and severe morbidity. Our goal was to test the hypothesis that an artificial intelligence (AI) based intervention can improve the diagnosis of pregnancy related cardiomyopathy.
Results: Among 1232 women randomized, 1195 completed baseline assessments (587 in the intervention arm and 608 in the control arm). The median age was 31 years and 39% were in their third trimester. AI-guided screening (digital stethoscope maximum prediction across all locations recorded) was associated with an increase in the diagnosis of cardiomyopathy - LVEF <50% (24/587 vs. 11/608; odds ratio 2.31, 95% CI: 1.12, 4.77; p=0.019). Among participants in the intervention arm, the digital stethoscope had an AUC of 0.95 (95% CI: 0.92, 0.99) for detection of LVEF <50% and 0.98 for detection of LVEF <40% (95% CI: 0.97, 0.99).
Conclusion: The study intervention resulted in double the number of cardiomyopathy cases diagnosed in the control arm, suggesting that half are likely under detected with usual care. In pregnant and postpartum women, AI-guided screening improves the diagnosis of pregnancy related cardiomyopathy, a potentially life-threatening and treatable condition.
MKT-0002785