AI Tool Speeds Echocardiogram Reporting With Comparable Accuracy

Echocardiography is the most widely used form of cardiac imaging, playing a key role in diagnosing heart disease and guiding treatment decisions. However, reviewing ultrasound images and documenting cardiac structure and function can be time-consuming, particularly as scan volumes increase. A new randomized study suggests that artificial intelligence-assisted echocardiogram reporting may reduce interpretation time while maintaining comparable accuracy.

Researchers within Kaiser Permanente Northern California evaluated an AI tool designed to analyze echocardiogram images and draft full preliminary reports. The system generated interpretations and measurements that cardiologists could review before finalizing the report. The study included 1,200 echocardiograms acquired during routine care over two weeks in March 2025. The exams came from hospital and outpatient clinic settings across 21 Kaiser Permanente Northern California medical centers.

The randomized trial compared preliminary reports prepared by the AI tool, sonographers, and cardiologists. Each group prepared a preliminary report for every echocardiogram. The reports were then randomly assigned to cardiologists, who finalized them without knowing whether the draft came from AI, a sonographer, or another cardiologist. Accuracy was assessed by how often cardiologists made substantial changes before finalizing the report.

Cardiologists made substantial changes to 50.6% of AI-generated preliminary interpretations. This compared with 49.7% of preliminary interpretations prepared by sonographers and cardiologists. The AI-assisted approach met the study criteria for noninferiority compared with human-assisted approaches. Cardiologists also spent an average of 4.6 minutes reviewing and finalizing AI-generated reports, compared with 8.5 minutes to generate reports without AI assistance.

The study population had an average age of 69 years. Women made up 47% of participants, and men made up 53%. Nine sonographers with an average of 7.2 years of experience and 20 cardiologists with an average of 11.7 years of experience participated. The findings were presented as late-breaking science at the American Heart Association’s Scientific Sessions 2026 in Chicago, and Journal of the American Heart Association was listed as journal information.

The study did not assess whether AI-assisted echocardiogram interpretation improves long-term patient outcomes. Cardiologists also knew they were participating in the study, which may have influenced work patterns. More research is needed to evaluate performance in larger-scale real-world clinical settings with higher exam volumes.

“AI cannot, however, replace the human team. The cardiologist remains ultimately responsible for interpreting the echocardiogram and certifying the accuracy of the final report, even when AI assists with the process. Moreover, the physician applies clinical judgment and knowledge of the patient to the final interpretation to develop a personalized care plan for each individual,” said Keane K. Lee, M.D., M.S., a cardiologist at Kaiser Permanente Santa Clara Homestead Medical Center in Santa Clara, California.

Related Links
Kaiser Permanente