Itzhak Zachi Attia, Ph.D., works in artificial intelligence, machine learning and signal processing. In his research, he uses machine learning to develop tools that enable the detection and prediction of diseases using cardiac biosignals.
Heart disease is the leading cause of death in the United States. But many people are diagnosed and treated only after symptoms appear, after developing severe morbidities or when they present with sudden cardiac death. People who have no symptoms manifest silent markers. These markers precede overt disease and are recorded during routine tests. But they are missed by clinicians, as these biomarkers are invisible to the human eye.
Dr. Attia uses multimodal cardiac data including electrocardiograms, echocardiograms and angiograms to develop artificial intelligence models that can detect treatable but silent diseases with high accuracy. Using pragmatic clinical trials, he collaborates closely with cardiologists, cardiac surgeons and other medical experts to test these models and incorporate them into clinical practice. The goal is to empower clinicians to apply these methods in practice without requiring any AI expertise.
- Disease detection. Dr. Attia and his colleagues develop neural network-based models to find patterns in biological signals by using millions of medical records with outcome labels. Critical biopatterns are detected without human supervision. This allows the networks to perform tasks that cannot be done by experts without the technology.
- Transformation of practice. Clinicians have better access to patient data with the tools that Dr. Attia and his colleagues create. The AI model results, including the AI dashboard that is embedded in the medical record, help facilitate ECG review and summarize AI-ECG model results graphically for clinical application.
- Digital clinical trials. While AI models are often very accurate in silico, they might be affected by real-life issues that are not modeled in the data used to train them, such as racial and sexual biases, data shifts, and low predictive power. Together with colleagues from Mayo Clinic's cardiology department, the Center for Digital Health and the Robert D. and Patricia E. Kern Center for the Science of Health Care Delivery, Dr. Attia participates in the design and implementation of digital, pragmatic clinical trials to test the effect of AI models on patient outcomes and patient care.
- Explainable AI. Most AI models today are considered black boxes. The signal features are detected without human supervision and are thus not known, and the models possess superhuman abilities. Dr. Attia and his colleagues aim to understand the drivers of the models, and the patterns they find, so that the data may lead to new biological insights, and increased trust in model results.
Significance to patient care
Studies conducted by Dr. Attia and his colleagues show the positive effects of the use of AI-ECG models in practice for both patients and clinicians. Patients are helped by the diagnosis of otherwise concealed, treatable disease. Clinicians can use these clinically integrated tools to find and treat diseases earlier, and without additional burden, allowing them to practice at the top of their license. For example, nurse practitioners can diagnose left ventricular dysfunction from a standard 12-lead ECG and arrange for cardiovascular referral. They can diagnose silent atrial fibrillation from a normal sinus rhythm ECG, and detect potentially life-threatening diseases such as ventricular dysfunction from patients' personal smartwatches or other mobile devices.