Key result
Data augmentation applied to a GE-based 3D-CNN model enabled accurate LVEF prediction across vendor machines, with mean absolute errors of 4.33 (GE), 4.42 (Philips), and 4.89 (Canon).
Why the study?
Medical AI applications in echocardiography face challenges from a lack of image data harmonization across different vendor machines.
Does data augmentation improve the accuracy of a 3D-CNN model trained on single-vendor echocardiographic data when predicting LVEF on multi-vendor data?
Does data augmentation improve the accuracy of a 3D-CNN model trained on single-vendor echocardiographic data when predicting LVEF on multi-vendor data?
Data augmentation techniques effectively harmonize echocardiographic images across different vendors, enabling accurate LVEF prediction by AI models trained on single-vendor data.
May broaden AI LVEF assessment across vendors without retraining; extends single-vendor models but leaves open prospective validation.
One of the common challenges in medical artificial intelligence (AI) applications using echocardiography is the lack of image data harmonization. This study aims to improve the prediction accuracy of left ventricular ejection fraction (LVEF) AI models by incorporating data augmentation (DA) techniques to address the variability in image data across different vendor machines. A database comprising 15,770 echocardiographic videos from 3154 patients across five different centers was utilized, with the data acquired using various vendor machines. We prepared datasets specific to GE healthcare (GE), philips (PH), and canon medical systems (CA) vendors, including 1911, 804, and 427 cohorts, respectively. A three-dimensional convolutional neural network (3D-CNN) was trained to predict LVEF, using videos consisting of 20 images per heartbeat as input, with training performed exclusively on data from GE machines (GE-based model). DA techniques, including gamma correction, scaling, median filtering, unsharp masking, translation, rotation, noise correction, and image conversion using a cycle generative adversarial network, were applied. A regression analysis using five different chamber views was performed on the GE test data. The accuracy of LVEF prediction using the GE-based model with DA-specifically gamma correction, scaling, and translation corrections-was excellent. The mean absolute error (root mean square error) in test cohorts was 4.33 (5.58) for GE, 4.42 (5.57) for PH, and 4.89 (6.57) for CA, which were comparable to the results of a model developed using data from all vendors, demonstrating the effectiveness of DA in harmonizing images. This study demonstrates that echocardiographic videos are transferable across vendors and that DA is highly effective in improving LVEF predictions from data acquired using different vendor machines.
No takes yet. Share an insight, caveat, or question.
Iwasaki et al. (2025) studied Left ventricular ejection fraction prediction (n=3,154). Data augmentation techniques (gamma correction, scaling, translation) on a 3D-CNN model vs. Model developed using data from all vendors was evaluated on Mean absolute error and root mean square error of LVEF prediction. Data augmentation applied to a GE-based 3D-CNN model enabled accurate LVEF prediction across vendor machines, with mean absolute errors of 4.33 (GE), 4.42 (Philips), and 4.89 (Canon).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: