Author: Amal Bilal, Class of 2028

Neurocardiology is a new and emerging field that examines the heart-brain interaction in the context of health and disease. Conditions such as stroke and cardiac arrhythmia are connected by the heart-brain axis: a network of neural, vascular, and physiological signals. Early and accurate detection of abnormalities along this axis is essential for improving patient outcomes. Stony Brook Medicine researcher Jade Basem and his colleagues evaluated the potential for artificial intelligence (AI) and machine learning (ML) to enhance diagnosis, prediction, and treatment of diseases that affect the nervous and cardiovascular systems.

The researchers applied ML models, including deep learning, convolutional neural networks, and gradient-boosting algorithms, to electrocardiograms (EKGs), echocardiograms, cardiac MRI, and CT scans. After being trained on the relevant data, these models were particularly accurate in diagnosing conditions such as atrial fibrillation, valvular heart disease, congenital defects, heart failure, and stroke risk prediction. In particular, the EKG model accurately detected atrial fibrillation with up to 90% accuracy, even in difficult cases, and deep learning analysis of echocardiograms had automatically identified early signs of heart failure or valve malfunctions. Similarly, AI-based MRI segmentation techniques produced accurate measurements of cardiac structure and function, drastically reducing diagnostic time and human error. The ability of AI to detect post-stroke arrhythmias or coronary artery disease, and diagnose and treat brain-to-heart disorders like Takotsubo cardiomyopathy, demonstrates the widespread universal applications of artificial intelligence in medical settings. In these cases, machine learning models have predicted mortality risk, identified imaging markers, and distinguished between disease types more accurately than conventional methods. Moreover, the researchers report that wearable technologies, such as AI-enhanced mobile EKGs and photoplethysmography sensors, have the potential to expand real-time cardiac monitoring to everyday settings, improving early detection and prevention of stroke.

However, despite the ongoing breakthroughs in AI, the researchers highlight several challenges that remain, such as the need for standardized data, clinician training, transparency within algorithms, and equitable access to technology. Furthermore, before AI is completely integrated into clinical neurocardiology, ethical concerns such as data privacy, bias in model design, and excessive dependence on automation need to be addressed. Finally, the study concludes that artificial intelligence and machine learning technologies are leading the way in the neurocardiology field. With increased diagnostic accuracy, personalized treatment, and prevention of stroke and cardiac disease, and continued collaboration between clinicians, engineers, and ethics specialists, AI tools will help expand the knowledge and treatment of heart and brain diseases.

Figure 1. A clinician interacts with a brain interface to visualize neural data to enhance diagnosis and treatment.

Works Cited:

[1] Basem, Jade et al. “Clinical applications of artificial intelligence and machine learning in neurocardiology: a comprehensive review.” Frontiers in cardiovascular medicine vol. 12 1525966. 3 Apr. 2025, doi:10.3389/fcvm.2025.1525966

[2] Image retrieved from: https://www.pexels.com/photo/close-up-shot-of-mri-results-4226264/