UltraSight Study: AI Guidance Enables Novices to Perform Diagnostic-Quality Cardiac Ultrasound After One Day of Training

A multicenter study published in European Heart Journal – Digital Health has found that novice operators, including nurses and medical students with no prior ultrasound experience, achieved diagnostic-quality cardiac ultrasound images in 97.7% of cases following eight hours of training with UltraSight’s AI-guided imaging system.

The prospective secondary analysis enrolled nine novice operators across three academic medical centers. Participants performed limited transthoracic echocardiography (TTE) scans on 159 patients after a single standardized training day. Three blinded expert cardiologists independently graded the acquired images for diagnostic quality.

Results were consistent across patient subgroups. Expert reviewers were able to rule out left ventricular dysfunction in 99.4% of cases and left ventricular hypertrophy in 98.7% of cases. All nine operators met the diagnostic-quality threshold from their very first independent scan.

“What stood out in this study was how quickly novice operators were able to achieve diagnostic-quality image acquisition across a range of patient types and clinical environments,”

Andrew Goldsmith, MD, MBA, Medical Director of UltraSight

The findings suggest AI-guided acquisition substantially reduces the training time traditionally required before novice users can perform echocardiograms independently.

The study comes amid well-documented shortages of cardiac sonographers and growing demand for bedside cardiac assessment in emergency and critical care settings, where timely imaging can directly affect triage and treatment decisions.

UltraSight’s system has received FDA 510(k) clearance. The company positions its platform as a tool to expand point-of-care cardiac imaging workflows under qualified physician oversight, rather than as a replacement for specialist sonographers.

The full study is available at https://doi.org/10.1093/ehjdh/ztag065.


Caroline Schissel, Daniel Trotzky, Noa Avisar, Talia Amar, Itay Kezurer, Dan Spiegelstein, Christopher W Baugh, Nicole M Duggan, Sachin Shah, Sarju Ganatra, Sourbha Dani, Andrew Goldsmith, Artificial intelligence in breaking the learning curve for echocardiography: a secondary analysis of a multicentre trial, European Heart Journal – Digital Health, Volume 7, Issue 4, May 2026, ztag065, https://doi.org/10.1093/ehjdh/ztag065