A deep learning pipeline for automated ventricular segmentation in single ventricle physiology showed minimal bias for end-diastolic (-0.6 mL/m2) and end-systolic (-1.1 mL/m2) volumes.
Observational (n=725)
Yes
Does an end-to-end deep learning pipeline accurately assess ventricular volumes from cardiac MRI in patients with single ventricle physiology compared to manual segmentation?
A deep learning pipeline can provide fast, standardized, and accurate automated ventricular segmentation for cardiac MRI in patients with complex single ventricle physiology across multiple centers.
Effect estimate: Bias: -0.6 mL/m2 (EDV), -1.1 mL/m2 (ESV) (95% CI LOA: -20.6 to 19.5 mL/m2 (EDV), -18.1 to 15.9 mL/m2 (ESV))
Purpose To develop an end-to-end deep learning (DL) pipeline for automated ventricular segmentation of cardiac MRI data from a multicenter registry of patients with Fontan circulation (Fontan Outcomes Registry Using CMR Examinations FORCE). Materials and Methods This retrospective study used 250 cardiac MRI examinations (November 2007–December 2022) from 13 institutions for training, validation, and testing. The pipeline contained three DL models: a classifier to identify short-axis cine stacks and two U-Net 3+ models for image cropping and segmentation. The automated segmentations were evaluated on the test set (n = 50) by using the Dice score. Volumetric and functional metrics derived from DL and ground truth manual segmentations were compared using Bland-Altman and intraclass correlation analysis. The pipeline was further qualitatively evaluated on 475 unseen examinations. Results There were acceptable limits of agreement (LOA) and minimal biases between the ground truth and DL end-diastolic volume (EDV) (bias: −0.6 mL/m2, LOA: −20.6 to 19.5 mL/m2) and end-systolic volume (ESV) (bias: −1.1 mL/m2, LOA: −18.1 to 15.9 mL/m2), with high intraclass correlation coefficients (ICCs > 0.97) and Dice scores (EDV, 0.91 and ESV, 0.86). There was moderate agreement for ventricular mass (bias: −1.9 g/m2, LOA: −17.3 to 13.5 g/m2) and an ICC of 0.94. There was also acceptable agreement for stroke volume (bias: 0.6 mL/m2, LOA: −17.2 to 18.3 mL/m2) and ejection fraction (bias: 0.6%, LOA: −12.2% to 13.4%), with high ICCs (>0.81). The pipeline achieved satisfactory segmentation in 68% of the 475 unseen examinations, while 26% needed minor adjustments, 5% needed major adjustments, and in 0.4%, the cropping model failed. Conclusion The DL pipeline can provide fast standardized segmentation for patients with single ventricle physiology across multiple centers. This pipeline can be applied to all cardiac MRI examinations in the FORCE registry. Keywords: Cardiac, Adults and Pediatrics, MR Imaging, Congenital, Volume Analysis, Segmentation, Quantification Supplemental material is available for this article. © RSNA, 2023
Yao et al. (Wed,) conducted a observational in Single ventricle physiology (Fontan circulation) (n=725). Deep learning pipeline for automated ventricular segmentation vs. Manual segmentation (ground truth) was evaluated on Volumetric and functional metrics (EDV, ESV) evaluated by Bland-Altman and intraclass correlation (Bias: -0.6 mL/m2 (EDV), -1.1 mL/m2 (ESV), 95% CI LOA: -20.6 to 19.5 mL/m2 (EDV), -18.1 to 15.9 mL/m2 (ESV)). A deep learning pipeline for automated ventricular segmentation in single ventricle physiology showed minimal bias for end-diastolic (-0.6 mL/m2) and end-systolic (-1.1 mL/m2) volumes.