Tilborghs Sofie, Liang Tiffany, Raptis Stavroula, Ishikita Ayako, Budts Werner, Dresselaers Tom, Bogaert Jan, Maes Frederik, Wald Rachel M, Van De Bruaene Alexander
Department of Electrical Engineering, Division of Processing Speech and Images (ESAT/PSI), KU Leuven, Leuven, Belgium; Medical Imaging Research Center, UZ Leuven, Leuven, Belgium.
Division of Cardiology, Peter Munk Cardiac Centre, University of Toronto, Toronto, Canada.
J Cardiovasc Magn Reson. 2024;26(2):101092. doi: 10.1016/j.jocmr.2024.101092. Epub 2024 Sep 11.
Deep learning is the state-of-the-art approach for automated segmentation of the left ventricle (LV) and right ventricle (RV) in cardiovascular magnetic resonance (CMR) images. However, these models have been mostly trained and validated using CMR datasets of structurally normal hearts or cases with acquired cardiac disease, and are therefore not well-suited to handle cases with congenital cardiac disease such as tetralogy of Fallot (TOF). We aimed to develop and validate a dedicated model with improved performance for LV and RV cavity and myocardium quantification in patients with repaired TOF.
We trained a three-dimensional (3D) convolutional neural network (CNN) with 5-fold cross-validation using manually delineated end-diastolic (ED) and end-systolic (ES) short-axis image stacks obtained from either a public dataset containing patients with no or acquired cardiac pathology (n = 100), an institutional dataset of TOF patients (n = 96), or both datasets mixed. Our method allows for missing labels in the training images to accommodate for different ED and ES phases for LV and RV as is commonly the case in TOF. The best performing model was applied to all frames of a separate test set of TOF cases (n = 36) and ED and ES phases were automatically determined for LV and RV separately. The model was evaluated against the performance of a commercial software (suiteHEART®, NeoSoft, Pewaukee, Wisconsin, US).
Training on the mixture of both datasets yielded the best agreement with the manual ground truth for the TOF cases, achieving a median Dice similarity coefficient of (93.8%, 89.8%) for LV cavity and of (92.9%, 90.9%) for RV cavity at (ED, ES) respectively, and of 80.9% and 61.8% for LV and RV myocardium at ED. The offset in automated ED and ES frame selection was 0.56 and 0.89 frames on average for LV and RV respectively. No statistically significant differences were found between our model and the commercial software for LV quantification (two-sided Wilcoxon signed rank test, p<5%), while RV quantification was significantly improved with our model achieving a mean absolute error of 12 ml for RV cavity compared to 36 ml for the commercial software.
We developed and validated a fully automatic segmentation and quantification approach for LV and RV, including RV mass, in patients with repaired TOF. Compared to a commercial software, our approach is superior for RV quantification indicating its potential in clinical practice.
深度学习是心血管磁共振（CMR）图像中左心室（LV）和右心室（RV）自动分割的先进方法。然而，这些模型大多使用结构正常心脏或获得性心脏病病例的CMR数据集进行训练和验证，因此不太适合处理法洛四联症（TOF）等先天性心脏病病例。我们旨在开发并验证一种专门的模型，以提高修复后TOF患者左心室和右心室腔及心肌定量的性能。
我们使用从包含无心脏病理或获得性心脏病理患者的公共数据集（n = 100）、TOF患者的机构数据集（n = 96）或两者混合的数据集中获得的手动勾勒的舒张末期（ED）和收缩末期（ES）短轴图像堆栈，通过5折交叉验证训练三维（3D）卷积神经网络（CNN）。我们的方法允许训练图像中存在缺失标签，以适应TOF中常见的左心室和右心室不同的ED和ES阶段。性能最佳的模型应用于TOF病例单独测试集的所有帧（n = 36），并分别自动确定左心室和右心室的ED和ES阶段。该模型与商业软件（suiteHEART®，NeoSoft，美国威斯康星州皮沃基）的性能进行了比较评估。
在两个数据集的混合数据上进行训练，对于TOF病例，与手动地面真值的一致性最佳，左心室腔在（ED，ES）时的中位骰子相似系数分别为（93.8%，89.8%），右心室腔为（92.9%，90.9%），舒张末期左心室和右心室心肌的相似系数分别为80.9%和61.8%。左心室和右心室自动选择ED和ES帧的偏移平均分别为0.56帧和0.89帧。在左心室定量方面，我们的模型与商业软件之间未发现统计学显著差异（双侧Wilcoxon符号秩检验，p<5%），而在右心室定量方面，我们的模型有显著改进，右心室腔的平均绝对误差为12 ml，而商业软件为36 ml。
我们开发并验证了一种用于修复后TOF患者左心室和右心室（包括右心室质量）的全自动分割和定量方法。与商业软件相比，我们的方法在右心室定量方面更具优势，表明其在临床实践中的潜力。
