Machine learning derived segmentation of phase velocity encoded cardiovascular magnetic resonance for fully automated aortic flow quantification

被引:57
作者
Bratt, Alex [1 ]
Kim, Jiwon [1 ,2 ]
Pollie, Meridith [2 ]
Beecy, Ashley N. [2 ]
Tehrani, Nathan H. [2 ]
Codella, Noel [3 ]
Perez-Johnston, Rocio [4 ]
Palumbo, Maria Chiara [2 ]
Alakbarli, Javid [2 ]
Colizza, Wayne [1 ]
Drexler, Ian R. [1 ]
Azevedo, Clerio F. [5 ]
Kim, Raymond J. [5 ]
Devereux, Richard B. [2 ]
Weinsaft, Jonathan W. [1 ,2 ,4 ,6 ]
机构
[1] Weill Cornell Med, Dept Radiol, 525 E 68th St, New York, NY 10065 USA
[2] Weill Cornell Med, Greenberg Div Cardiol, Dept Med, 525 E 68th St, New York, NY 10065 USA
[3] IBM TJ Watson Res Ctr, 1101 Kitchawan Rd, Yorktown Hts, NY 10598 USA
[4] Mem Sloan Kettering Canc Ctr, 1275 York Ave, New York, NY 10065 USA
[5] Duke Cardiovasc Magnet Resonance Ctr, 10 Duke Med Circle, Durham, NC 27710 USA
[6] Weill Cornell Med Coll, 525 East 68th St, New York, NY 10021 USA
基金
美国国家卫生研究院;
关键词
Cardiovascular magnetic resonance; Machine learning; Deep learning; Phase contrast; Aorta; VALIDATION; ALGORITHM; MRI;
D O I
10.1186/s12968-018-0509-0
中图分类号
R5 [内科学];
学科分类号
1002 ; 100201 ;
摘要
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 <0.01min/case (1.2min for entire dataset); manual segmentation required 3.960.36min/case (12.5h 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 +/- 3.18mL, p<0.01): Nearly all (98%) of cases differed by 5mL 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 +/- 17.7mL, p=0.36). Among advanced mitral regurgitation patients, machine learning yielded lower stroke volume than did volumetric cine-CMR ( 12.6 +/- 20.9mL, 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 +/- 1.77mL, 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 +/- 2.25mL, p=0.25).Conclusion p id=Par4 Fully 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.
引用
收藏
页数:11
相关论文
共 30 条
  • [1] Bai W, 2017, ARXIV171009289CS
  • [2] An Exploration of 2D and 3D Deep Learning Techniques for Cardiac MR Image Segmentation
    Baumgartner, Christian F.
    Koch, Lisa M.
    Pollefeys, Marc
    Konukoglu, Ender
    [J]. STATISTICAL ATLASES AND COMPUTATIONAL MODELS OF THE HEART: ACDC AND MMWHS CHALLENGES, 2018, 10663 : 111 - 119
  • [3] Beerbaum P, 2001, CIRCULATION, V103, P2476
  • [4] Deep Learning Techniques for Automatic MRI Cardiac Multi-Structures Segmentation and Diagnosis: Is the Problem Solved?
    Bernard, Olivier
    Lalande, Alain
    Zotti, Clement
    Cervenansky, Frederick
    Yang, Xin
    Heng, Pheng-Ann
    Cetin, Irem
    Lekadir, Karim
    Camara, Oscar
    Gonzalez Ballester, Miguel Angel
    Sanroma, Gerard
    Napel, Sandy
    Petersen, Steffen
    Tziritas, Georgios
    Grinias, Elias
    Khened, Mahendra
    Kollerathu, Varghese Alex
    Krishnamurthi, Ganapathy
    Rohe, Marc-Michel
    Pennec, Xavier
    Sermesant, Maxime
    Isensee, Fabian
    Jaeger, Paul
    Maier-Hein, Klaus H.
    Full, Peter M.
    Wolf, Ivo
    Engelhardt, Sandy
    Baumgartner, Christian F.
    Koch, Lisa M.
    Wolterink, Jelmer M.
    Isgum, Ivana
    Jang, Yeonggul
    Hong, Yoonmi
    Patravali, Jay
    Jain, Shubham
    Humbert, Olivier
    Jodoin, Pierre-Marc
    [J]. IEEE TRANSACTIONS ON MEDICAL IMAGING, 2018, 37 (11) : 2514 - 2525
  • [5] STATISTICAL METHODS FOR ASSESSING AGREEMENT BETWEEN TWO METHODS OF CLINICAL MEASUREMENT
    BLAND, JM
    ALTMAN, DG
    [J]. LANCET, 1986, 1 (8476) : 307 - 310
  • [6] Practical value of cardiac magnetic resonance imaging for clinical quantification of aortic valve stenosis comparison with echocardiography
    Caruthers, SD
    Lin, SJ
    Brown, P
    Watkins, MP
    Williams, TA
    Lehr, KA
    Wickline, SA
    [J]. CIRCULATION, 2003, 108 (18) : 2236 - 2243
  • [7] VoxResNet: Deep voxelwise residual networks for brain segmentation from 3D MR images
    Chen, Hao
    Dou, Qi
    Yu, Lequan
    Qin, Jing
    Heng, Pheng-Ann
    [J]. NEUROIMAGE, 2018, 170 : 446 - 455
  • [8] Automatic MR Prostate Segmentation by Deep Learning with Holistically-Nested Networks
    Cheng, Ruida
    Roth, Holger R.
    Lay, Nathan
    Lu, Le
    Turkbey, Baris
    Gandler, William
    McCreedy, Evan S.
    Choyke, Peter
    Summers, Ronald M.
    McAuliffe, Matthew J.
    [J]. MEDICAL IMAGING 2017: IMAGE PROCESSING, 2017, 10133
  • [9] Improved Left Ventricular Mass Quantification With Partial Voxel Interpolation In Vivo and Necropsy Validation of a Novel Cardiac MRI Segmentation Algorithm
    Codella, Noel C. F.
    Lee, Hae Yeoun
    Fieno, David S.
    Chen, Debbie W.
    Hurtado-Rua, Sandra
    Kochar, Minisha
    Finn, John Paul
    Judd, Robert
    Goyal, Parag
    Schenendorf, Jesse
    Cham, Matthew D.
    Devereux, Richard B.
    Prince, Martin
    Wang, Yi
    Weinsaft, Jonathan W.
    [J]. CIRCULATION-CARDIOVASCULAR IMAGING, 2012, 5 (01) : 137 - 146
  • [10] Rapid and Accurate Left Ventricular Chamber Quantification Using a Novel CMR Segmentation Algorithm: A Clinical Validation Study
    Codella, Noel C. F.
    Cham, Matthew D.
    Wong, Richard
    Chu, Christopher
    Min, James K.
    Prince, Martin R.
    Wang, Yi
    Weinsaft, Jonathan W.
    [J]. JOURNAL OF MAGNETIC RESONANCE IMAGING, 2010, 31 (04) : 845 - 853