Artificial intelligence (AI) has shown promising results in assessing heart health in various studies and applications. AI algorithms can analyze large amounts of medical data and detect patterns that may be missed by human doctors. Additionally, AI can help to automate some tasks, such as reading electrocardiograms (ECGs) or echocardiograms, which can save time and improve accuracy.
When compared to echocardiogram assessments performed by sonographers, researchers discovered AI to be superior in assessing and diagnosing cardiac function. After reading an echocardiogram, who can better assess and diagnose cardiac function: artificial intelligence (AI) or a sonographer?
According to Cedars-Sinai researchers and their research published today in the peer-reviewed journal Nature, AI outperformed echocardiogram assessments performed by sonographers in assessing and diagnosing cardiac function.
The findings are the result of a first-of-its-kind, blinded, randomized clinical trial of AI in cardiology led by Cedars-Sinai researchers from the Smidt Heart Institute and the Division of Artificial Intelligence in Medicine.
This successful clinical trial sets a superb precedent for how novel clinical AI algorithms can be discovered and tested within health systems, increasing the likelihood of seamless deployment for improved patient care.
Sumeet Chugh
“The results have immediate implications for patients undergoing cardiac function imaging as well as broader implications for the field of cardiac imaging,” said cardiologist David Ouyang, MD, the clinical trial’s principal investigator and senior author. “This trial provides rigorous evidence that using AI in this novel way can improve the quality and effectiveness of echocardiogram imaging for a large number of patients.”
The researchers are confident that this technology will be beneficial when implemented across the clinical system at Cedars-Sinai and in health systems across the country.
“This successful clinical trial sets a superb precedent for how novel clinical AI algorithms can be discovered and tested within health systems, increasing the likelihood of seamless deployment for improved patient care,” said Sumeet Chugh, MD, director of the Division of Artificial Intelligence in Medicine and Pauline and Harold Price Chair in Cardiac Electrophysiology Research.
In 2020, Smidt Heart Institute and Stanford University researchers created one of the first AI technologies to assess cardiac function, specifically left ventricular ejection fraction – the key heart measurement used in diagnosing cardiac function. Their findings were also published in Nature.
Based on those findings, the new study investigated whether AI was more accurate in evaluating 3,495 transthoracic echocardiogram studies by comparing initial assessment by AI or by a sonographer – also known as an ultrasound technician.
- Cardiologists more frequently agreed with the AI initial assessment and made corrections to only 16.8% of the initial assessments made by AI.
- Cardiologists made corrections to 27.2% of the initial assessments made by sonographers.
- The physicians were unable to tell which assessments were made by AI and which were made by sonographers.
“We asked our cardiologists to guess whether the preliminary interpretation was performed by AI or by a sonographer, and it turns out that they couldn’t tell the difference,” Ouyang explained. “This demonstrates the AI algorithm’s strong performance as well as its seamless integration into clinical software.” We believe that these are all positive indicators for future AI trial research in the field.”
The goal, according to Ouyang, is to save clinicians time and reduce the more time-consuming aspects of the cardiac imaging workflow. However, the cardiologist remains the final expert adjudicator of the AI model output. The clinical trial, as well as subsequent published research, shed light on the possibility of regulatory approvals.
“This work raises the bar for artificial intelligence technologies being considered for regulatory approval, as the Food and Drug Administration has previously approved artificial intelligence tools without data from prospective clinical trials,” said Susan Cheng, MD, MPH, director of the Institute for Research on Healthy Aging in the Department of Cardiology at the Smidt Heart Institute and study co-senior author. “We believe that this level of evidence provides clinicians with additional assurance as health systems work to more broadly adopt artificial intelligence as part of efforts to increase overall efficiency and quality.”
