Deep learning to estimate cardiac magnetic resonance-derived left ventricular mass.
Deep learning to estimate cardiac magnetic resonance-derived left ventricular mass.
复制标题
DOI：
10.1016/j.cvdhj.2021.03.001
复制
发表时间：
2021-04
影响因子：
--
通讯作者：
Lubitz SA
中科院分区：
文献类型：
--
作者：
Khurshid S;Friedman SF;Pirruccello JP;Di Achille P;Diamant N;Anderson CD;Ellinor PT;Batra P;Ho JE;Philippakis AA;Lubitz SA
心脏磁共振(CMR)是诊断左心室肥厚(LVH)的金标准。可以使用专有算法(如InlineVF)估计CMR得出的LV质量，但其准确性和可用性可能会受到限制。开发一个开源的深度学习模型来估计CMR衍生的LV质量。在接受CMR的英国生物库前瞻性队列的参与者中，我们训练了两个卷积神经网络来估计左心室质量。第一个(ML4Hreg)通过手动标记的LV团块(在5065人中可用)进行回归，而第二个(ML4Hseg)通过InlineVF(版本D13A)等高线进行LV分割。我们使用皮尔逊相关性和平均绝对误差(MAE)将ML4Hreg、ML4Hseg和InlineVF与独立抵抗集中手动标记的LV质量块进行了比较。我们通过对年龄和性别进行调整后的Logistic回归分析，评估了CMR所致的左心室肥厚与常见心血管疾病之间的关系。我们在38,574名个体中生成了基于CMR的LV质量估计。与ML4Hreg(r=0.843，95%CI 0.823~0.861；MAE10.51，95%CI 9.86~11.15，P=0.01)和InlineVF(r=0.795，95%CI 0.770~0.818)相比，ML4Hseg能更准确地再现手工标记的左心室包块(r=0.864，95%可信区间0.847~0.880；MAE 10.41g，95%CI 9.82~10.99)；MAE 14.30，95%可信区间13.46-11.01，P<；0.01)。用ML4Hseg定义的左心室肥厚与高血压(优势比2.76，95%可信区间2.51~3.04)、房颤(1.75，95%可信区间1.37~2.20)和心力衰竭(4.67，95%可信区间3.28~6.49)的相关性最强。ML4Hseg是一个开源的深度学习模型，提供CMR衍生的左心室质量的自动量化。以心脏结构为特征的深度学习模型可能有助于更广泛的心血管发现。
Cardiac magnetic resonance (CMR) is the gold standard for left ventricular hypertrophy (LVH) diagnosis. CMR-derived LV mass can be estimated using proprietary algorithms (eg, InlineVF), but their accuracy and availability may be limited. To develop an open-source deep learning model to estimate CMR-derived LV mass. Within participants of the UK Biobank prospective cohort undergoing CMR, we trained 2 convolutional neural networks to estimate LV mass. The first (ML4Hreg) performed regression informed by manually labeled LV mass (available in 5065 individuals), while the second (ML4Hseg) performed LV segmentation informed by InlineVF (version D13A) contours. We compared ML4Hreg, ML4Hseg, and InlineVF against manually labeled LV mass within an independent holdout set using Pearson correlation and mean absolute error (MAE). We assessed associations between CMR-derived LVH and prevalent cardiovascular disease using logistic regression adjusted for age and sex. We generated CMR-derived LV mass estimates within 38,574 individuals. Among 891 individuals in the holdout set, ML4Hseg reproduced manually labeled LV mass more accurately (r = 0.864, 95% confidence interval [CI] 0.847–0.880; MAE 10.41 g, 95% CI 9.82–10.99) than ML4Hreg (r = 0.843, 95% CI 0.823–0.861; MAE 10.51, 95% CI 9.86–11.15, P = .01) and InlineVF (r = 0.795, 95% CI 0.770–0.818; MAE 14.30, 95% CI 13.46–11.01, P < .01). LVH defined using ML4Hseg demonstrated the strongest associations with hypertension (odds ratio 2.76, 95% CI 2.51–3.04), atrial fibrillation (1.75, 95% CI 1.37–2.20), and heart failure (4.67, 95% CI 3.28–6.49). ML4Hseg is an open-source deep learning model providing automated quantification of CMR-derived LV mass. Deep learning models characterizing cardiac structure may facilitate broad cardiovascular discovery.
登录
查看更多内容
影响因子：
19.7
作者：
Kawel-Boehm, Nadine;Kronmal, Richard;Bluemke, David A.
通讯作者：
Bluemke, David A.
影响因子：
37.8
作者：
Aung, Nay;Vargas, Jose D.;Petersen, Steffen E.
通讯作者：
Petersen, Steffen E.
DOI：
10.1016/s0735-1097(98)00407-0
发表时间：
1998-11-01
影响因子：
24
作者：
Haider, AW;Larson, MG;Levy, D
通讯作者：
Levy, D
影响因子：
39.3
作者：
Littlejohns TJ;Sudlow C;Allen NE;Collins R
通讯作者：
Collins R
影响因子：
15.8
作者：
Sudlow C;Gallacher J;Allen N;Beral V;Burton P;Danesh J;Downey P;Elliott P;Green J;Landray M;Liu B;Matthews P;Ong G;Pell J;Silman A;Young A;Sprosen T;Peakman T;Collins R
通讯作者：
Collins R