Development and testing of a deep learning-based strategy for scar segmentation on CMR-LGE images

ObjectiveThe aim of this paper is to investigate the use of fully convolutional neural networks (FCNNs) to segment scar tissue in the left ventricle from cardiac magnetic resonance with late gadolinium enhancement (CMR-LGE) images.MethodsA successful FCNN in the literature (the ENet) was modified and trained to provide scar-tissue segmentation. Two segmentation protocols (Protocol 1 and Protocol 2) were investigated, the latter limiting the scar-segmentation search area to the left ventricular myocardial tissue region. CMR-LGE from 30 patients with ischemic-heart disease were retrospectively analyzed, for a total of 250 images, presenting high variability in terms of scar dimension and location. Segmentation results were assessed against manual scar-tissue tracing using one-patient-out cross validation.ResultsProtocol 2 outperformed Protocol 1 significantly (p value < 0.05), with median sensitivity and Dice similarity coefficient equal to 88.07% [inter-quartile range (IQR) 18.84%] and 71.25% (IQR 31.82%), respectively.DiscussionBoth segmentation protocols were able to detect scar tissues in the CMR-LGE images but higher performance was achieved when limiting the search area to the myocardial region. The findings of this paper represent an encouraging starting point for the use of FCNNs for the segmentation of nonviable scar tissue from CMR-LGE images.

[1]  L. R. Dice.  Measures of the Amount of Ecologic Association Between Species , 1945 .

[2]  Katherine C. Wu,et al.  Accuracy of Contrast-Enhanced Magnetic Resonance Imaging in Predicting Improvement of Regional Myocardial Function in Patients After Acute Myocardial Infarction , 2002, Circulation.

[3]  Alejandro F. Frangi,et al.  Evaluation of state-of-the-art segmentation algorithms for left ventricle infarct from late Gadolinium enhancement MR images , 2016, Medical Image Anal..

[4]  Sebastian Kelle,et al.  Prognostic value of myocardial infarct size and contractile reserve using magnetic resonance imaging. , 2009, Journal of the American College of Cardiology.

[5]  G. Kitagawa,et al.  Akaike Information Criterion Statistics , 1988 .

[6]  Jürgen Schmidhuber,et al.  Deep learning in neural networks: An overview , 2014, Neural Networks.

[7]  Sebastian Ruder,et al.  An overview of gradient descent optimization algorithms , 2016, Vestnik komp'iuternykh i informatsionnykh tekhnologii.

[8]  Scott D Flamm,et al.  Standardized image interpretation and post processing in cardiovascular magnetic resonance: Society for Cardiovascular Magnetic Resonance (SCMR) Board of Trustees Task Force on Standardized Post Processing , 2013, Journal of Cardiovascular Magnetic Resonance.

[9]  Tianqi Chen,et al.  Empirical Evaluation of Rectified Activations in Convolutional Network , 2015, ArXiv.

[10]  Daniel Rueckert,et al.  Medical Image Computing and Computer-Assisted Intervention − MICCAI 2017: 20th International Conference, Quebec City, QC, Canada, September 11-13, 2017, Proceedings, Part II , 2017, Lecture Notes in Computer Science.

[11]  Simon Just Kjeldgaard Pedersen.  Circular Hough Transform , 2009, Encyclopedia of Biometrics.

[12]  Wail Gueaieb,et al.  3D scar segmentation from LGE-MRI using a continuous max-flow method , 2018, Medical Imaging.

[13]  Guang Yang,et al.  Multiview Two-Task Recursive Attention Model for Left Atrium and Atrial Scars Segmentation , 2018, MICCAI.

[14]  Gunnar Seemann,et al.  Functional Imaging and Modeling of the Heart , 2003, Lecture Notes in Computer Science.

[15]  Felix Lau,et al.  FastVentricle: Cardiac Segmentation with ENet , 2017, FIMH.

[16]  Russell H. Taylor,et al.  Surgical data science for next-generation interventions , 2017, Nature Biomedical Engineering.

[17]  Sergey Ioffe,et al.  Rethinking the Inception Architecture for Computer Vision , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[18]  Wail Gueaieb,et al.  Comparison of myocardial scar geometries from 2D and 3D LGE-MRI , 2018, Medical Imaging.

[19]  Vicente Grau,et al.  Automated LGE Myocardial Scar Segmentation Using MaskSLIC Supervoxels - Replicating the Clinical Method , 2017, MIUA.

[20]  Vivek Muthurangu,et al.  Evaluation of techniques for the quantification of myocardial scar of differing etiology using cardiac magnetic resonance. , 2011, JACC. Cardiovascular imaging.

[21]  Eranga Ukwatta,et al.  Myocardial scar segmentation from magnetic resonance images using convolutional neural network , 2018, Medical Imaging.

[22]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[23]  Nicolai Schoch,et al.  Surgical Data Science: Enabling Next-Generation Surgery , 2017, ArXiv.

[24]  R. Kim,et al.  Contrast-enhanced magnetic resonance imaging of myocardium at risk: distinction between reversible and irreversible injury throughout infarct healing. , 2000, Journal of the American College of Cardiology.

[25]  William M. Wells,et al.  Simultaneous truth and performance level estimation (STAPLE): an algorithm for the validation of image segmentation , 2004, IEEE Transactions on Medical Imaging.

[26]  Heinz-Otto Peitgen,et al.  Mixture-Model-Based Segmentation of Myocardial Delayed Enhancement MRI , 2012, STACOM.

[27]  Guang Yang,et al.  Fully automatic segmentation and objective assessment of atrial scars for long‐standing persistent atrial fibrillation patients using late gadolinium‐enhanced MRI , 2017, Medical physics.

[28]  Elena De Momi,et al.  Blood vessel segmentation algorithms - Review of methods, datasets and evaluation metrics , 2018, Comput. Methods Programs Biomed..

[29]  Dan W Rettmann,et al.  Accurate and Objective Infarct Sizing by Contrast-enhanced Magnetic Resonance Imaging in a Canine Myocardial Infarction Model , 2022 .

[30]  Eric Saloux,et al.  Scar extent evaluated by late gadolinium enhancement CMR: a powerful predictor of long term appropriate ICD therapy in patients with coronary artery disease , 2013, Journal of Cardiovascular Magnetic Resonance.

[31]  Thomas O'Donnell,et al.  Quantification of Delayed Enhancement MR Images , 2004, MICCAI.

[32]  P. Kellman,et al.  Quantitative myocardial infarction on delayed enhancement MRI. Part I: Animal validation of an automated feature analysis and combined thresholding infarct sizing algorithm , 2006, Journal of magnetic resonance imaging : JMRI.

[33]  Jian Sun,et al.  Identity Mappings in Deep Residual Networks , 2016, ECCV.

[34]  Eugenio Culurciello,et al.  ENet: A Deep Neural Network Architecture for Real-Time Semantic Segmentation , 2016, ArXiv.

[35]  Felix Lau,et al.  ScarGAN: Chained Generative Adversarial Networks to Simulate Pathological Tissue on Cardiovascular MR Scans , 2018, DLMIA/ML-CDS@MICCAI.

[36]  Xun Liu,et al.  Quantized Segmentation of Fibrotic Tissue of Left Atrial from Delay-Enhancement MRI Images Using Level-Set and Graph-Cut , 2017, 2017 International Conference on Machine Vision and Information Technology (CMVIT).

[37]  Seyed-Ahmad Ahmadi,et al.  V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation , 2016, 2016 Fourth International Conference on 3D Vision (3DV).

[38]  Richard D. White,et al.  Quantitative assessment of myocardial scar in delayed enhancement magnetic resonance imaging , 2003, Journal of magnetic resonance imaging : JMRI.

[39]  Pierre Croisille,et al.  Comparison of visual scoring and quantitative planimetry methods for estimation of global infarct size on delayed enhanced cardiac MRI and validation with myocardial enzymes. , 2011, European journal of radiology.

[40]  Guang Yang,et al.  Segmenting Atrial Fibrosis from Late Gadolinium-Enhanced Cardiac MRI by Deep-Learned Features with Stacked Sparse Auto-Encoders , 2017, MIUA.

[41]  Gerhard Adam,et al.  Acute myocardial infarction: evaluation with first-pass enhancement and delayed enhancement MR imaging compared with 201Tl SPECT imaging. , 2004, Radiology.

[42]  Mihaela Pop,et al.  Quantification of fibrosis in infarcted swine hearts by ex vivo late gadolinium-enhancement and diffusion-weighted MRI methods , 2013, Physics in medicine and biology.

[43]  Heinz-Otto Peitgen,et al.  A Comprehensive Approach to the Analysis of Contrast Enhanced Cardiac MR Images , 2008, IEEE Transactions on Medical Imaging.

[44]  Zhengqin Li,et al.  Superpixel segmentation using Linear Spectral Clustering , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[45]  O. Simonetti,et al.  The use of contrast-enhanced magnetic resonance imaging to identify reversible myocardial dysfunction. , 2000, The New England journal of medicine.

[46]  S. Vrieze.  Model selection and psychological theory: a discussion of the differences between the Akaike information criterion (AIC) and the Bayesian information criterion (BIC). , 2012, Psychological methods.

[47]  G. Kitagawa,et al.  Akaike Information Criterion Statistics , 1988 .

[48]  Mauro Pepi,et al.  Comparison of Image Processing Techniques for Nonviable Tissue Quantification in Late Gadolinium Enhancement Cardiac Magnetic Resonance Images , 2016, Journal of thoracic imaging.