Paul C. Lauterbur Research Center for Biomedical Imaging, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, SZ University Town, Shenzhen, 518055, China.
CAS Key Lab of Molecular Imaging, Institute of Automation, Chinese Academy of Sciences, Beijing, 100190, China.
Eur Radiol. 2021 Jun;31(6):3931-3940. doi: 10.1007/s00330-020-07454-9. Epub 2020 Nov 25.
The high variability of hypertrophic cardiomyopathy (HCM) genetic phenotypes has prompted the establishment of risk-stratification systems that predict the risk of a positive genetic mutation based on clinical and echocardiographic profiles. This study aims to improve mutation-risk prediction by extracting cardiovascular magnetic resonance (CMR) morphological features using a deep learning algorithm.
We recruited 198 HCM patients (48% men, aged 47 ± 13 years) and divided them into training (147 cases) and test (51 cases) sets based on different genetic testing institutions and CMR scan dates (2012, 2013, respectively). All patients underwent CMR examinations, HCM genetic testing, and an assessment of established genotype scores (Mayo Clinic score I, Mayo Clinic score II, and Toronto score). A deep learning (DL) model was developed to classify the HCM genotypes, based on a nonenhanced four-chamber view of cine images.
The areas under the curve (AUCs) for the test set were Mayo Clinic score I (AUC: 0.64, sensitivity: 64.29%, specificity: 47.83%), Mayo Clinic score II (AUC: 0.70, sensitivity: 64.29%, specificity: 65.22%), Toronto score (AUC: 0.74, sensitivity: 75.00%, specificity: 56.52%), and DL model (AUC: 0.80, sensitivity: 85.71%, specificity: 69.57%). The combination of the DL and the Toronto score resulted in a significantly higher predictive performance (AUC = 0.84, sensitivity: 83.33%, specificity: 78.26%), compared with Mayo I (p = 006), Mayo II (p = 022), and Toronto score (p = 0.029).
The combination of the DL model, based on nonenhanced cine CMR images and the Toronto score yielded significantly higher diagnostic performance in detecting HCM mutations.
• Deep learning method could enable the extraction of image features from cine images. • Deep learning method based on cine images performed better than established scores in identifying HCM patients with positive genotypes. • The combination of the deep learning method based on cine images and the Toronto score could further improve the performance of the identification of HCM patients with positive genotypes.
肥厚型心肌病（HCM）遗传表型的高度变异性促使建立了风险分层系统，该系统基于临床和超声心动图特征预测阳性基因突变的风险。本研究旨在通过使用深度学习算法提取心血管磁共振（CMR）形态特征来改善突变风险预测。
我们招募了 198 名 HCM 患者（48%为男性，年龄 47±13 岁），并根据不同的基因检测机构和 CMR 扫描日期（分别为 2012 年和 2013 年）将其分为训练（147 例）和测试（51 例）集。所有患者均接受 CMR 检查、HCM 基因检测和已建立的基因型评分（Mayo 诊所评分 I、Mayo 诊所评分 II 和多伦多评分）评估。基于电影图像的非增强四腔视图，开发了一种深度学习（DL）模型来对 HCM 基因型进行分类。
测试集的曲线下面积（AUC）分别为 Mayo 诊所评分 I（AUC：0.64，灵敏度：64.29%，特异性：47.83%）、Mayo 诊所评分 II（AUC：0.70，灵敏度：64.29%，特异性：65.22%）、多伦多评分（AUC：0.74，灵敏度：75.00%，特异性：56.52%）和 DL 模型（AUC：0.80，灵敏度：85.71%，特异性：69.57%）。DL 模型与多伦多评分的结合显著提高了预测性能（AUC=0.84，灵敏度：83.33%，特异性：78.26%），与 Mayo I（p=0.06）、Mayo II（p=0.02）和多伦多评分（p=0.029）相比。
基于非增强电影 CMR 图像和多伦多评分的 DL 模型组合在检测 HCM 突变方面具有更高的诊断性能。
深度学习方法可以从电影图像中提取图像特征。
基于电影图像的深度学习方法在识别具有阳性基因型的 HCM 患者方面优于已建立的评分。
基于电影图像的深度学习方法与多伦多评分的结合可以进一步提高识别具有阳性基因型的 HCM 患者的性能。
