Researchers at the Mount Sinai Health System have developed an innovative generative AI transformer model that improves the accuracy and effectiveness of ECG (electrocardiogram)-related diagnoses.
In a study published in the June 6 online issue of npj Digital Medicine, the team reported that its new deep-learning model, HeartBEiT, surpassed established commonly used AI methods for ECG analysis — convolutional neural networks (CNNs), used for computer-vision tasks.
Researchers pre-trained HeartBEiT on 8.5 million ECGs from 2.1 million patients collected over four decades from four hospitals within the Mount Sinai Health System. Then they tested the model's performance against CNN architectures commonly used in cardiac diagnostic areas.
The study found that HeartBEiT had "significantly higher performance at lower sample sizes, along with better 'explainability' and can perform as well as, if not better than, these methods, using a tenth of the data," according to Akhil Vaid, MD, Instructor of Data-Driven and Digital Medicine (D3M) at the Icahn School of Medicine at Mount Sinai.
The researchers tested the model on three tasks: learning if a patient is having a heart attack; if they have a genetic disorder called hypertrophic cardiomyopathy; and how effectively their heart is functioning. "In each case, our model performed better than all other tested baselines.”
This study was funded by the National Heart, Lung, and Blood Institute of the NIH and by the National Center for Advancing Translational Sciences of the NIH.
Citation: Vaid, A., Jiang, J., Sawant, A., Lerakis, S., Argulian, E., Ahuja, Y., Lampert, J., Charney, A., Greenspan, H., Narula, J., Glicksberg, B., & Nadkarni, G. N. (2023). A foundational vision transformer improves diagnostic performance for electrocardiograms. Npj Digital Medicine, 6(1), 1-8. https://doi.org/10.1038/s41746-023-00840-9 (open-access)