Automatic segmentation of myocardium at risk from contrast enhanced SSFP CMR: validation against expert readers and SPECT.
Automatic segmentation of myocardium at risk from contrast enhanced SSFP CMR: validation against expert readers and SPECT.
复制标题
DOI：
10.1186/s12880-016-0124-1
复制
发表时间：
2016-03-05
影响因子：
2.7
通讯作者：
Heiberg E
中科院分区：
文献类型：
--
作者：
Tufvesson J;Carlsson M;Aletras AH;Engblom H;Deux JF;Koul S;Sörensson P;Pernow J;Atar D;Erlinge D;Arheden H;Heiberg E
再灌注治疗的疗效可以通过确定危险心肌（MaR）和心肌梗死（MI）的大小来评估心肌挽救指数（MSI），（MSI = 1-MI/MaR）。心血管磁共振 (CMR) 可用于通过晚期钆增强 (LGE) 评估 MI，通过 T2 加权成像或对比增强 SSFP (CE-SSFP) 评估 MaR。 LGE 针对 MI 以及 T2 加权成像针对 MaR 开发并验证了自动分割算法。然而，没有可用于 CE-SSFP 的算法。因此，本研究的目的是开发和验证 CE-SSFP 中 MaR 的自动分割。自动算法应用表面线圈强度校正，并通过期望最大化对心肌强度进行分类，以根据先验区域标准定义 MaR 区域，并根据 LGE 定义梗塞区域。自动分割与专家读者对来自两项多中心随机临床试验 (RCT)（CHILL-MI 和 MITOCARE）的 183 名再灌注急性 MI 患者的手动描绘进行了验证，并与另一组中的心肌灌注 SPECT 进行了验证（n = 16）。在舒张末期和收缩末期手动描绘心内膜和心外膜边界。将 MaR 的手动勾画用作参考，并评估一部分患者 (n = 15) 中 MaR 的手动勾画和自动分割的观察者间变异性。 MaR 表示为左心室质量 (%LVM) 的百分比，并通过偏倚进行分析（平均值±标准差）。通过骰子相似系数（DSC）（平均值±标准差）分析区域一致性。通过手动和自动分割评估的 MaR 分别为 36±±10% 和 37±±11%LVM，偏差为 1±±6%LVM，区域一致性 DSC 0.85±±0.08 (n=±183)。 SPECT和CE-SSFP自动分割评估的MaR分别为27±±10%LVM和29±±7%LVM，偏差为2±±7%LVM。手动描绘的观察者间变异性为 0±±3%LVM，自动分割的观察者间变异性为 -1±±2%LVM。 CE-SSFP 中 MaR 的自动分割与多中心、多供应商研究中的手动划分进行了验证，具有低偏差和高区域一致性。偏差和变异性与手动描绘的观察者间变异性相似，并且自动分割减少了观察者间变异性。因此，所提出的自动分割可用于减少 RCT 中 MaR 量化的主观性。 NCT01379261。 NCT01374321。本文的在线版本 (doi:10.1186/s12880-016-0124-1) 包含补充材料，可供授权用户使用。
Efficacy of reperfusion therapy can be assessed as myocardial salvage index (MSI) by determining the size of myocardium at risk (MaR) and myocardial infarction (MI), (MSI = 1-MI/MaR). Cardiovascular magnetic resonance (CMR) can be used to assess MI by late gadolinium enhancement (LGE) and MaR by either T2-weighted imaging or contrast enhanced SSFP (CE-SSFP). Automatic segmentation algorithms have been developed and validated for MI by LGE as well as for MaR by T2-weighted imaging. There are, however, no algorithms available for CE-SSFP. Therefore, the aim of this study was to develop and validate automatic segmentation of MaR in CE-SSFP. The automatic algorithm applies surface coil intensity correction and classifies myocardial intensities by Expectation Maximization to define a MaR region based on a priori regional criteria, and infarct region from LGE. Automatic segmentation was validated against manual delineation by expert readers in 183 patients with reperfused acute MI from two multi-center randomized clinical trials (RCT) (CHILL-MI and MITOCARE) and against myocardial perfusion SPECT in an additional set (n = 16). Endocardial and epicardial borders were manually delineated at end-diastole and end-systole. Manual delineation of MaR was used as reference and inter-observer variability was assessed for both manual delineation and automatic segmentation of MaR in a subset of patients (n = 15). MaR was expressed as percent of left ventricular mass (%LVM) and analyzed by bias (mean ± standard deviation). Regional agreement was analyzed by Dice Similarity Coefficient (DSC) (mean ± standard deviation). MaR assessed by manual and automatic segmentation were 36 ± 10 % and 37 ± 11 %LVM respectively with bias 1 ± 6 %LVM and regional agreement DSC 0.85 ± 0.08 (n = 183). MaR assessed by SPECT and CE-SSFP automatic segmentation were 27 ± 10 %LVM and 29 ± 7 %LVM respectively with bias 2 ± 7 %LVM. Inter-observer variability was 0 ± 3 %LVM for manual delineation and -1 ± 2 %LVM for automatic segmentation. Automatic segmentation of MaR in CE-SSFP was validated against manual delineation in multi-center, multi-vendor studies with low bias and high regional agreement. Bias and variability was similar to inter-observer variability of manual delineation and inter-observer variability was decreased by automatic segmentation. Thus, the proposed automatic segmentation can be used to reduce subjectivity in quantification of MaR in RCT. NCT01379261. NCT01374321. The online version of this article (doi:10.1186/s12880-016-0124-1) contains supplementary material, which is available to authorized users.