Related Experiment Videos
Multi-task artificial intelligence annotation of echocardiographic images: a retrospective multi-cohort study
Yuki Sahashi1, David Choi1, Hirotaka Ieki2,3
1Department of Cardiology, Smidt Heart Institute, Cedars-Sinai Medical Center, Los Angeles, CA.
Insights
EchoNet-Segmentation is a new AI tool that automates key echocardiogram measurements, improving accuracy and efficiency for cardiac assessments. This open-source framework supports clinical workflows by providing generalizable deep learning models for VTI and atrial area.
Area of Science:
- Artificial Intelligence in Medical Imaging
- Cardiovascular Ultrasound Technology
- Deep Learning for Echocardiography
Background:
- Transthoracic echocardiograms require extensive manual annotation of over 70 parameters, leading to significant workload and inter-observer variability.
- Existing open-source models primarily focus on 2D B-mode ventricular function, neglecting spectral Doppler and atrial measurements crucial for valvular and diastolic assessments.
Purpose of the Study:
- To develop and validate EchoNet-Segmentation, a comprehensive suite of deep learning models for automated segmentation and measurement of left and right atrial area and velocity-time integral (VTI) in echocardiograms.
- To provide an open-source framework addressing the gap in automated Doppler and atrial measurements for echocardiographic analysis.
Main Methods:
- Retrospective multi-cohort study utilizing 186,712 sonographer-annotated echocardiogram images from 93,978 studies (56,855 patients) for training.
- Development of task-specific deep learning segmentation models for atrial area and VTI measurements.
- Performance evaluation on internal held-out test sets, a temporal split cohort, an external Kaiser Permanente Northern California cohort, and the public MIMIC-Echo dataset.
Main Results:
- EchoNet-Segmentation demonstrated strong agreement with sonographer measurements on the internal test set (VTI R² 0.817-0.882, atrial area R² 0.675-0.747).
- Consistent performance was observed across diverse datasets, including an external cohort (VTI R² 0.575-0.859, atrial area R² 0.803-0.876) and on different vendor machines.
- The AI models outperformed a general-purpose medical image foundation model (MedSAM2) in cardiac chamber segmentation tasks.
Conclusions:
- EchoNet-Segmentation is the first open-source framework offering accurate and generalizable automated measurements for key echocardiographic parameters, facilitating end-to-end automation.
- The public release of model weights, code, and demonstration tools promotes reproducibility, research, and clinical deployment of AI in echocardiography.
- Further prospective validation is necessary to ascertain the impact of automated measurements on diagnostic accuracy, workflow efficiency, and patient outcomes.
Background:
A comprehensive transthoracic echocardiogram involves the assessment of over 70 parameters, placing a substantial burden on sonographers and physicians for manual annotation with considerable inter-observer variability. Prior open-source segmentation models have largely addressed 2D B-mode ventricular function, leaving a gap in the spectral Doppler and atrial measurements required for valvular and diastolic assessment such as velocity-time integral (VTI) and atrial chamber size.
Methods:
In this retrospective multi-cohort study, we developed EchoNet-Segmentation, comprehensive task-specific deep learning segmentation models for left and right atrial area and VTI Doppler measurements. Training used 186,712 sonographer-annotated images from 93,978 studies (56,855 patients) at Cedars-Sinai Medical Center (CSMC). Performance was evaluated on a held-out CSMC test set, a CSMC temporal split, an external Kaiser Permanente Northern California cohort, and the public MIMIC-Echo dataset.
Findings:
On the CSMC held-out test set, our AI models showed strong agreement with sonographer measurements, with R2 of 0.817-0.882 and mean absolute error (MAE) of 1.13-3.80 cm for automated VTI measurements, and R2 of 0.675-0.747 and MAE of 2.48-2.52 cm2 for left and right atrial area segmentation. Performance was consistently confirmed on the CSMC temporal split (VTI: R2 0.606-0.866, atrial area: R2 0.694-0.705) and on the KPNC external cohort (VTI: R2 0.575-0.859, atrial area: R2 0.803-0.876), on the MIMIC-Echo dataset. Robustness was demonstrated on a different vendor's machines and across subgroups. EchoNet-Segmentation outperformed an open-source medical image foundation model with bounding-box, point prompt configurations on R2, MAE, and Dice score on both held-out test dataset and MIMIC apical four-chamber data.
Interpretation:
EchoNet-Segmentation is the first open-source framework that delivers accurate, generalizable automated measurement across several key routine echocardiographic parameters, supporting end-to-end automation of clinically important echocardiographic assessments. Public release of model weights, code, and demonstration tools can facilitate reproducibility, research use and clinical deployment.
Related Concept Videos
01:17Imaging Studies for Cardiovascular System I:Echocardiography
Indications: Echocardiography is utilized to diagnose heart failure, valve disorders, and myocardial infarction. It also assesses cardiac structures' size, shape, and motion, evaluates...
01:20Imaging Studies for Cardiovascular System II:Types of Echocardiography
Types of Echocardiography
Transthoracic Echocardiography (TTE)
TTE is the most common type of echocardiogram which involves placing a transducer on the patient's chest, emitting sound waves to create heart images. TTE is invaluable for evaluating the heart's size, structure, and motion, making it particularly useful for diagnosing...