STRESS echocardiography (SE) is a key diagnostic tool for coronary artery disease (CAD), but its accuracy varies significantly based on the clinician’s expertise. The PROTEUS trial aimed to evaluate the effectiveness of an AI tool, EchoGo Pro, in aiding clinical decision-making during SE for patients with suspected CAD. Specifically, the research team sought to determine whether AI could standardise decision-making across clinicians of varying experience levels.
The trial was conducted across 20 UK hospitals, enrolling 2,341 patients (mean age 64 years; 45% female; 20% with pre-existing heart disease) with suspected CAD between November 2021 and June 2023. Participants were randomly assigned to either standard clinical decision-making or AI-augmented decision-making using EchoGo Pro in a 1:1 ratio. The primary outcome measured was the appropriateness of decisions to refer patients for invasive coronary angiograms and the occurrence of acute coronary events within six months.
The trial found no significant differences between AI-assisted and standard decision-making in correctly selecting patients for angiography. Among those sent for angiography, correct referrals were slightly higher in the AI group (34 out of 49) compared to the control group (27 out of 36), but this was not statistically significant. However, AI showed potential benefits for less experienced clinicians and in clinically complex cases, suggesting it could help standardise accuracy across different experience levels.
While the PROTEUS trial did not demonstrate significant overall improvements in decision-making with AI assistance, it highlighted the potential for AI to enhance decision-making among less experienced clinicians and in complex cases. The findings suggest that AI could help equalize the accuracy of diagnoses across clinicians, regardless of their experience level.
Abigail Craig, EMJ
Reference
Woodward G et al. PROTEUS study: A prospective randomized controlled trial evaluating the use of artificial intelligence in stress echocardiography. Am Heart J. 2023;263:123-32.
