- Us study shows AI tool twice as likely to diagnose heart condition early
- researchers have found that clinicians who most frequently followed the recommendations of an AI tool were twice as likely to diagnose low ejection fraction (EF)
Artificial intelligence is poised to transform medicine and according to a new study by the US-based Mayo Clinic, researchers have found that clinicians who most frequently followed the recommendations of an AI tool were twice as likely to diagnose low ejection fraction (EF), an indication that the heart is not functioning as well as it should. Early diagnosis and treatment in patients with low EF are vital to reduce the risk of heart failure.
Few studies have looked at the characteristics of clinicians who have quickly adopted AI tools (high adopters) versus those who are more hesitant (low adopters), as well as the clinical outcomes associated with these two approaches. According to the study, which was published last week in Mayo Clinic Proceedings, the proportion of AI-positive people with confirmed low EF was 33.9% in high adopters and 16.3% in low adopters.
The study’s lead researcher, Dr. David Rushlow of the Department of Family Medicine at the Mayo Clinic in Rochester, USA, stated that the team’s goal was to compare the characteristics of “high adopters” and “low adopters” of an artificial intelligence (AI)-enabled electrocardiogram (ECG) algorithm that alerted for possible low left ventricular ejection fraction (EF) and the subsequent effectiveness of detecting patients with low EF.
Prof K Srinath Reddy, who formerly headed the department of cardiology at All India Institute of Medical Sciences, New Delhi, and is founder-president of Public Health Foundation of India, said AI can help collate large data sets and distill them into diagnostic pattern recognition and management algorithms. It can thereby overcome gaps in prior clinical experience. This can be very helpful in promoting early recognition, prompt care, and appropriate referral practices in primary care. “However, the algorithms developed by AI are dependent on the representativeness and accuracy of the input data. They are context-dependent and algorithms developed in Western populations are not automatically always applicable in the Indian context. So, we need to develop our own AI capabilities, using large Indian data sets. It is clear, however, that those who can adapt well to the use of AI can improve their diagnostic accuracy and provide better patient care. This is especially so in primary care settings where care providers do not have specialist training,” he said.
Dr. Sanjeev Jadhav, Director of the Heart Lung Transplant program at Apollo Hospital, Navi Mumbai, said that technology has helped in diagnosing and treating patients in various fields of medicine and surgery. “AI is still at a basic level and such tools generated in the context of the Mayo Clinic study will evolve soon to help doctors diagnose right at the primary level whether the patients will have a heart problem or whether there is a problem with the heart function. Most patients approach the doctor when they have symptoms like heaviness in the chest or difficulty in breathing – AI will diagnose heart conditions before the symptoms start prompting early treatment. Technology will help doctors diagnose patients way before symptoms start and this will help in the future,” Dr. Jadhav said.
The study included a total of 165 clinicians and 11,573 patients. Electronic health records were used to collect clinical data. If a patient was 18 years of age or older and had received an ECG for any reason between August 5, 2019, and March 31, 2020, their data were included in the analysis. During the study period, the decision to order an echocardiogram was based solely on an individual patient’s first ECG. Data from patients were excluded if they knew their EF was less than or equal to 50% or if they had a history of heart failure before the ECG. The study found that the proportion of AI-positive with confirmed low EF was 33 .9 percent for high adopters and 16.3 percent for low adopters.
AI tools, according to researchers, must be adopted by clinicians to have an impact on human health. The study’s researchers discovered a wide range in the rate of adoption of AI recommendations. Those who responded to the AI and ordered an echocardiogram on their patients were significantly more likely to detect left ventricular dysfunction (33.9 percent vs 16.3 percent).
For decades, AI has promised to augment clinician decision-making in health care, according to researchers. Recent advancements, such as the adoption of electronic health records (EHRs) and the ability to apply machine learning to this massive data repository, have enabled them to use AI to improve diagnostic accuracy and refine treatment plans. In the study, they discovered that primary care clinicians who were early adopters of an AI-enabled clinical decision support tool were twice as likely to diagnose low EF as those who were late adopters.
They also concluded that clinicians who were most likely to follow the AI decision aid’s recommendations had less experience dealing with complex patients, emphasizing the importance of clinician education and engagement, as well as AI systems that integrate seamlessly into the workflows of busy caregivers.
