A deep learning neural network significantly reduced the mean absolute error for left atrial volume estimation compared to the area-length method (3.5 mL vs 13.0 mL, P<0.05).
Does a deep learning neural network improve the accuracy of estimating 3D left atrial shape, volume, and surface area from 2CH and 4CH views compared to the bi-plane area-length method?
A deep learning 3D UNet significantly improves the accuracy of estimating left atrial volume and surface area from standard 2D views compared to the traditional bi-plane area-length method.
Absolute Event Rate: 3.5% vs 13%
p-value: p=< 0.05
AIMS: Left atrial volume is commonly estimated using the bi-plane area-length method from two-chamber (2CH) and four-chamber (4CH) long axes views. However, this can be inaccurate due to a violation of geometric assumptions. We aimed to develop a deep learning neural network to infer 3D left atrial shape, volume and surface area from 2CH and 4CH views. METHODS AND RESULTS: A 3D UNet was trained and tested using 2CH and 4CH segmentations generated from 3D coronary computed tomography angiography (CCTA) segmentations (n = 1700, with 1400/100/200 cases for training/validating/testing). An independent test dataset from another institution was also evaluated, using cardiac magnetic resonance (CMR) 2CH and 4CH segmentations as input and 3D CCTA segmentations as the ground truth (n = 20). For the 200 test cases generated from CCTA, the network achieved a mean Dice score value of 93.7%, showing excellent 3D shape reconstruction from two views compared with the 3D segmentation Dice of 97.4%. The network also showed significantly lower mean absolute error values of 3.5 mL/4.9 cm2 for LA volume/surface area respectively compared to the area-length method errors of 13.0 mL/34.1 cm2 respectively (P < 0.05 for both). For the independent CMR test set, the network achieved accurate 3D shape estimation (mean Dice score value of 87.4%), and a mean absolute error values of 6.0 mL/5.7 cm2 for left atrial volume/surface area respectively, significantly less than the area-length method errors of 14.2 mL/19.3 cm2 respectively (P < 0.05 for both). CONCLUSIONS: Compared to the bi-plane area-length method, the network showed higher accuracy and robustness for both volume and surface area.
Xu et al. (Thu,) conducted a other in Left atrial volume and shape estimation (n=1,720). Deep learning neural network (3D UNet) vs. Bi-plane area-length method was evaluated on Mean absolute error for left atrial volume (p=< 0.05). A deep learning neural network significantly reduced the mean absolute error for left atrial volume estimation compared to the area-length method (3.5 mL vs 13.0 mL, P<0.05).