One of the most important means of preventing and treating cardiovascular disease is an electrocardiograph (ECG). As artificial intelligence (AI) becomes a much more advanced and reliable tool, the scope of ECG applications including diagnostic and predictive functions is expanding fast. A key paper in The Lancet Digital Health recently reported on an AI-enhanced ECG platform, called the AI-ECG risk estimator (AIRE), with impressive predictive power for death and cardiovascular risks.
The AIRE platform is a fully functional, readable, and biologically plausible AI-ECG risk-sensing platform that can be deployed for risk assessment in various clinical settings across the world and potentially contribute to cardiovascular disease prevention and treatment.
Researchers built the AIRE platform to combat issues with existing AI-ECG models used in clinical settings, such as their individual tractability, interpretability and biological plausibleness.
Based on patient-specific survival curves created from individual ECGs, AIRE can predict a patient's risk of death and accurately estimate the time to death, which was not possible with previous AI-ECG models.
Fig. 1. Eight models in the AIRE platform, training datasets, and validation datasets. (Sau A.; et al. 2024)
The AIRE platform was crafted by researchers based on deep learning and discrete-time survival modeling on the massive BIDMC dataset. The dataset contains 1,163,401 ECG data from 189,539 patients, providing a solid foundation for AIRE's accurate prediction.
In addition to its accurate predictive capability, AIRE is also robustly interpretive and biologically plausible. Through phenotype-wide association studies and genome-wide association studies, the researchers revealed the underlying biological mechanisms by which AIRE predicts increased risk, including changes in cardiac structure and function, as well as genes associated with cardiac structure, biological aging, and metabolic syndrome.
To validate the effectiveness of AIRE, the research team selected five different multinational cohorts from the United States, Brazil and the United Kingdom for validation, including volunteers, primary care patients and secondary care patients. These cohorts cover a wide range of patient populations, making the validation results of the AIRE platform broadly representative and applicable.
The results showed that AIRE performed well in predicting all-cause mortality, ventricular arrhythmias, atherosclerotic cardiovascular disease, and heart failure, with high concordance indexes. This result undoubtedly provides strong evidence for the clinical application of AIRE.
AIRE is a revolution in the field of ECG analysis, and it's not just a simple diagnostic tool. AIRE is also a platform for future health status of patients and decisions made by clinicians. As AI technology develops and clinical implementation deepens, AIRE will soon be the gold standard of ECG analysis and it promises to transform the health care and treatment of patients as never before.
Protheragen-ING AI-Pharma is a leading AI technology company, dedicated to taking the enormous capabilities of AI and putting them to use, particularly in healthcare. And with our AI solutions, we can help medical organizations pinpoint diagnoses more accurately, offer more tailored treatments, and ultimately achieve better patient outcomes. We're excited to work with clinicians around the world to drive healthcare innovation and growth. If you are interested in our AI technology, please feel free to contact us.
Original Article:
Sau A.; et al. (2024). Artificial intelligence-enabled electrocardiogram for mortality and cardiovascular risk estimation: a model development and validation study. The Lancet Digital Health. 2024, 6(11): e791-e802.
Services Related in the Article:
Our mission is to accelerate the development of life-saving drugs by leveraging cutting-edge AI technologies.
Certification
