Automated right ventricular assessment in pediatric echocardiography via deep learning improves measurement reliability and reduces variability

Right ventricular (RV) function is important for pediatric cardiac evaluation but accurate and reproducible quantification of RV function is challenging. This study aimed to develop a deep learning (DL) model for RV functional assessment from echocardiography (ECHO) which out-performs current, manual methods. We trained multiple DL segmentation models, using a dataset of 664 pediatric ECHOs, and proceeded with the best performing model for evaluation. DL model performance was assessed using the dice similarity coefficient (DSC) for segmentation, mean absolute error (MAE) for RVFAC. Blinded expert evaluation was conducted between ground truth and model generated segmentation outputs. A detailed analysis of inter-observer variability identified the main sources of RVFAC variability among four experts and the DL model, as well as opportunities for the model to improve RV assessment in practice. The FCBFormer architecture yielded the best segmentation quality with DSC of 0.926 and MAE of 5.913% for RVFAC prediction. Blinded expert review revealed that model generated segmentation was favored over human in 57.3% of evaluated cases. All sources of variation were overcome by the RVFAC model: RV contour delineation, RV cardiac cycle selection, and RV end-diastolic/end-systolic frame identification. This study demonstrates the feasibility of DL-based automated RV functional assessment for pediatric patients, offering a promising approach for more consistent and systematic longitudinal tracking of RV function than manual ECHO assessment. • This work automates right ventricular assessment from 2D pediatric echocardiography, achieving overall higher segmentation performance than previously reported, with additional blinded clinical assessment of segmentation quality. In addition, the right ventricular segmented area is transformed into a metric of right ventricular function (fractional area change, RVFAC) with lower inter-observer variability than human annotation. Automatic computation of right ventricular function from pediatric echocardiography presents the opportunity for more reliable, automatic, and continuous assessment across all heartbeats acquired within a single scan. Furthermore, the reduction in inter-observer variability will enable more meaningful assessments of right ventricular function between patients and of the same patient across time.