BackgroundPhase contrast (PC) cardiovascular magnetic resonance (CMR) is widely employed for flow quantification, but analysis typically requires time consuming manual segmentation which can require human correction. Advances in machine learning have markedly improved automated processing, but have yet to be applied to PC-CMR. This study tested a novel machine learning model for fully automated analysis of PC-CMR aortic flow.MethodsA machine learning model was designed to track aortic valve borders based on neural network approaches. The model was trained in a derivation cohort encompassing 150 patients who underwent clinical PC-CMR then compared to manual and commercially-available automated segmentation in a prospective validation cohort. Further validation testing was performed in an external cohort acquired from a different site/CMR vendor.ResultsAmong 190 coronary artery disease patients prospectively undergoing CMR on commercial scanners (84% 1.5T, 16% 3T), machine learning segmentation was uniformly successful, requiring no human intervention: Segmentation time was < 0.01 min/case (1.2 min for entire dataset); manual segmentation required 3.96 ± 0.36 min/case (12.5 h for entire dataset). Correlations between machine learning and manual segmentation-derived flow approached unity (r = 0.99, p < 0.001). Machine learning yielded smaller absolute differences with manual segmentation than did commercial automation (1.85 ± 1.80 vs. 3.33 ± 3.18 mL, p < 0.01): Nearly all (98%) of cases differed by ≤5 mL between machine learning and manual methods. Among patients without advanced mitral regurgitation, machine learning correlated well (r = 0.63, p < 0.001) and yielded small differences with cine-CMR stroke volume (∆ 1.3 ± 17.7 mL, p = 0.36). Among advanced mitral regurgitation patients, machine learning yielded lower stroke volume than did volumetric cine-CMR (∆ 12.6 ± 20.9 mL, p = 0.005), further supporting validity of this method. Among the external validation cohort (n = 80) acquired using a different CMR vendor, the algorithm yielded equivalently small differences (∆ 1.39 ± 1.77 mL, p = 0.4) and high correlations (r = 0.99, p < 0.001) with manual segmentation, including similar results in 20 patients with bicuspid or stenotic aortic valve pathology (∆ 1.71 ± 2.25 mL, p = 0.25).ConclusionFully automated machine learning PC-CMR segmentation performs robustly for aortic flow quantification - yielding rapid segmentation, small differences with manual segmentation, and identification of differential forward/left ventricular volumetric stroke volume in context of concomitant mitral regurgitation. Findings support use of machine learning for analysis of large scale CMR datasets.Electronic supplementary materialThe online version of this article (10.1186/s12968-018-0509-0) contains supplementary material, which is available to authorized users.
Background
Intraoperative or post procedure right ventricular (RV) dysfunction confers a poor prognosis in the post-operative period. Conventional predictors for RV function are limited due the effect of cardiac surgery on traditional RV indices; novel echocardiographic techniques hold the promise to improve RV functional stratification.
Methods
Comprehensive echocardiographic data were collected prospectively during elective cardiac surgery. Tricuspid annular plane systolic excursion (TAPSE), peak RV systolic velocity (S′), and RV fractional area change (FAC) were quantified on transesophageal echo (TEE). RV global and regional (septal and free wall) longitudinal strain was quantified using speckle-tracking echo in RV-focused views. Two intraoperative time points were used for comparison: pre-sternotomy (baseline) and after chest closure.
Results
The population was comprised of 53 patients undergoing cardiac surgery [15.1% coronary artery bypass graft (CABG) only, 28.3% valve only, 50.9% combination (e.g. valve/CABG, valve/aortic graft) surgeries], among whom 38% had impaired RV function at baseline defined as RV FAC < 35%. All conventional RV functional indices including TAPSE, S′ and FAC declined immediately following CPB (1.5 ± 0.3 vs.1.1 ± 0.3 cm, 8.0 ± 2.1 vs. 6.2 ± 2.5 cm/s, 36.8 ± 9.3 vs. 29.3 ± 10.6%;
p
< 0.001 for all). However, left ventricular (LV) and RV hemodynamic parameters remained unchanged (LV ejection fraction (EF): 56.8 ± 13.0 vs. 55.8 ± 12.9%;
p
= 0.40, pulmonary artery systolic pressure (PASP): 26.5 ± 7.4 vs 27.3 ± 6.7 mmHg;
p
= 0.13). Speckle tracking echocardiographic data demonstrated a significant decline in RV global longitudinal strain (GLS) [19.0 ± 6.5 vs. 13.5 ± 6.9%,
p
< 0.001]. Pre-procedure FAC, GLS and free wall strain predicted RV dysfunction at chest closure (34.7 ± 9.1 vs. 41.6 ± 8.1%,
p
= 0.01, 17.7 ± 6.5 vs. 21.8 ± 5.4%;
p
= 0.03, 20.3 ± 6.4 vs. 24.2 ± 5.8%;
p
= 0.04), whereas traditional linear RV indices such as TAPSE and RV S′ at baseline had no impact on intraoperative RV dysfunction (
p
= NS for both).
Conclusions
Global and regional RV function, as measured by 2D indices and strain, acutely decline intraoperatively. Impaired RV strain is associated with intraoperative RV functional decline and provides incremental value to traditional RV indices in predicting those who will develop RV dysfunction.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.