Deep neural network ensemble for on-the-fly quality control-driven segmentation of cardiac MRI T1 mapping

Highlights • Quality control-driven framework for cardiac segmentation and quality control.• Exploiting variability within deep neural network ensemble to estimate uncertainty.• Novel on-the-fly selection mechanism for the final optimal segmentation.• Accurate, reliable, and fully automated analysis of T1 map with visualization.• Highlighting a potential flaw of the Pearson correlation to evaluate quality score.

[1]  Richard B. Thompson,et al.  Clinical recommendations for cardiovascular magnetic resonance mapping of T1, T2, T2* and extracellular volume: A consensus statement by the Society for Cardiovascular Magnetic Resonance (SCMR) endorsed by the European Association for Cardiovascular Imaging (EACVI) , 2017, Journal of Cardiovascular Magnetic Resonance.

[2]  Ender Konukoglu,et al.  PHiSeg: Capturing Uncertainty in Medical Image Segmentation , 2019, MICCAI.

[3]  Michael Brady,et al.  Deep Quantitative Liver Segmentation and Vessel Exclusion to Assist in Liver Assessment , 2017, MIUA.

[4]  van Hc Hans Assen,et al.  LV Challenge LKEB Contribution: Fully Automated Myocardial Contour Detection , 2009, The MIDAS Journal.

[5]  Jeonghwan Gwak,et al.  Ensemble of Instance Segmentation Models for Polyp Segmentation in Colonoscopy Images , 2019, IEEE Access.

[6]  Lin Yang,et al.  A New Ensemble Learning Framework for 3D Biomedical Image Segmentation , 2018, AAAI.

[7]  Klaus H. Maier-Hein,et al.  A Probabilistic U-Net for Segmentation of Ambiguous Images , 2018, NeurIPS.

[8]  Stefan Neubauer,et al.  Myocardial Tissue Characterization by Magnetic Resonance Imaging , 2014, Journal of thoracic imaging.

[9]  Andrew S Flett,et al.  Human non-contrast T1 values and correlation with histology in diffuse fibrosis , 2013, Heart.

[10]  Reza Nezafat,et al.  Automated analysis of cardiovascular magnetic resonance myocardial native T1 mapping images using fully convolutional neural networks , 2019, Journal of Cardiovascular Magnetic Resonance.

[11]  Ben Glocker,et al.  Ensemble of Convolutional Neural Networks Improves Automated Segmentation of Acute Ischemic Lesions Using Multiparametric Diffusion-Weighted MRI , 2019, American Journal of Neuroradiology.

[12]  William T. Clarke,et al.  Relationship Between Left Ventricular Structural and Metabolic Remodeling in Type 2 Diabetes , 2015, Diabetes.

[13]  Looi Chow Lee,et al.  Segmentation of the Left Ventricle from Cine MR Images Using a Comprehensive Approach , 2009, The MIDAS Journal.

[14]  Konrad Werys,et al.  Quality Control-Driven Image Segmentation Towards Reliable Automatic Image Analysis in Large-Scale Cardiovascular Magnetic Resonance Aortic Cine Imaging , 2019, MICCAI.

[15]  Ben Glocker,et al.  Automated quality control in image segmentation: application to the UK Biobank cardiovascular magnetic resonance imaging study , 2019, Journal of Cardiovascular Magnetic Resonance.

[16]  Suyash P. Awate,et al.  A Bayesian Neural Net to Segment Images with Uncertainty Estimates and Good Calibration , 2019, IPMI.

[17]  J. Zamorano,et al.  Innovative imaging methods in heart failure: a shifting paradigm in cardiac assessment. Position statement on behalf of the Heart Failure Association of the European Society of Cardiology , 2018, European journal of heart failure.

[18]  M. Robson,et al.  Pheochromocytoma Is Characterized by Catecholamine-Mediated Myocarditis, Focal and Diffuse Myocardial Fibrosis, and Myocardial Dysfunction. , 2016, Journal of the American College of Cardiology.

[19]  S. Piechnik,et al.  HIV-1–Related Cardiovascular Disease Is Associated With Chronic Inflammation, Frequent Pericardial Effusions, and Probable Myocardial Edema , 2016, Circulation. Cardiovascular imaging.

[20]  Stefan Neubauer,et al.  Shortened Modified Look-Locker Inversion recovery (ShMOLLI) for clinical myocardial T1-mapping at 1.5 and 3 T within a 9 heartbeat breathhold , 2010, Journal of cardiovascular magnetic resonance : official journal of the Society for Cardiovascular Magnetic Resonance.

[21]  Timo Kohlberger,et al.  Evaluating Segmentation Error without Ground Truth , 2012, MICCAI.

[22]  P. Matthews,et al.  Diffuse Myocardial Fibrosis and Inflammation in Rheumatoid Arthritis: Insights From CMR T1 Mapping. , 2015, JACC. Cardiovascular imaging.

[23]  G. Wright,et al.  Evaluation Framework for Algorithms Segmenting Short Axis Cardiac MRI. , 2009, The MIDAS Journal.

[24]  J. Francis,et al.  Adenosine stress native T1 mapping in severe aortic stenosis: evidence for a role of the intravascular compartment on myocardial T1 values , 2014, Journal of Cardiovascular Magnetic Resonance.

[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]  M. Robson,et al.  Noncontrast T1 mapping for the diagnosis of cardiac amyloidosis. , 2013, JACC. Cardiovascular imaging.

[27]  Ben Glocker,et al.  Automated cardiovascular magnetic resonance image analysis with fully convolutional networks , 2017, Journal of Cardiovascular Magnetic Resonance.

[28]  M. Robson,et al.  Native T1-mapping detects the location, extent and patterns of acute myocarditis without the need for gadolinium contrast agents , 2014, Journal of Cardiovascular Magnetic Resonance.

[29]  Ling Shao,et al.  A review of heart chamber segmentation for structural and functional analysis using cardiac magnetic resonance imaging , 2016, Magnetic Resonance Materials in Physics, Biology and Medicine.

[30]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

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

[32]  N. Geller,et al.  Hypertrophic Cardiomyopathy Registry: The rationale and design of an international, observational study of hypertrophic cardiomyopathy. , 2015, American heart journal.

[33]  Ben Glocker,et al.  Subject-level Prediction of Segmentation Failure using Real-Time Convolutional Neural Nets , 2018 .

[34]  P. Matthews,et al.  Subclinical myocardial inflammation and diffuse fibrosis are common in systemic sclerosis – a clinical study using myocardial T1-mapping and extracellular volume quantification , 2014, Journal of Cardiovascular Magnetic Resonance.

[35]  Xin Yang,et al.  Deep Learning Techniques for Automatic MRI Cardiac Multi-Structures Segmentation and Diagnosis: Is the Problem Solved? , 2018, IEEE Transactions on Medical Imaging.

[36]  Xiang Li,et al.  Estimating the ground truth from multiple individual segmentations incorporating prior pattern analysis with application to skin lesion segmentation , 2011, 2011 IEEE International Symposium on Biomedical Imaging: From Nano to Macro.

[37]  J. Pousin,et al.  Evaluation of the Dynamic Deformable Elastic Template model for the segmentation of the heart in MRI sequences , 2009, The MIDAS Journal.

[38]  Demetri Terzopoulos,et al.  Snakes: Active contour models , 2004, International Journal of Computer Vision.

[39]  M. Robson,et al.  Cardiovascular magnetic resonance by non contrast T1-mapping allows assessment of severity of injury in acute myocardial infarction , 2012, Journal of Cardiovascular Magnetic Resonance.

[40]  Sébastien Ourselin,et al.  STEPS: Similarity and Truth Estimation for Propagated Segmentations and its application to hippocampal segmentation and brain parcelation , 2013, Medical Image Anal..

[41]  Ben Glocker,et al.  Automated Quality Control in Image Segmentation: Application to the UK Biobank Cardiac MR Imaging Study , 2019, ArXiv.

[42]  Mert R. Sabuncu,et al.  Multi-atlas segmentation of biomedical images: A survey , 2014, Medical Image Anal..

[43]  Caroline Petitjean,et al.  A review of segmentation methods in short axis cardiac MR images , 2011, Medical Image Anal..

[44]  Charles Blundell,et al.  Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles , 2016, NIPS.

[45]  Matthew D. Robson,et al.  Systolic ShMOLLI myocardial T1-mapping for improved robustness to partial-volume effects and applications in tachyarrhythmias , 2015, Journal of Cardiovascular Magnetic Resonance.

[46]  P. Matthews,et al.  Imaging in population science: cardiovascular magnetic resonance in 100,000 participants of UK Biobank - rationale, challenges and approaches , 2013, Journal of Cardiovascular Magnetic Resonance.

[47]  Konstantinos Kamnitsas,et al.  Reverse Classification Accuracy: Predicting Segmentation Performance in the Absence of Ground Truth , 2017, IEEE Transactions on Medical Imaging.

[48]  M. Robson,et al.  Myocardial T1 mapping and extracellular volume quantification: a Society for Cardiovascular Magnetic Resonance (SCMR) and CMR Working Group of the European Society of Cardiology consensus statement , 2013, Journal of Cardiovascular Magnetic Resonance.

[49]  Stefan Neubauer,et al.  Myocardial Tissue Characterization Using Magnetic Resonance Noncontrast T1 Mapping in Hypertrophic and Dilated Cardiomyopathy , 2012, Circulation. Cardiovascular imaging.

[50]  M. Robson,et al.  Non-contrast T1-mapping detects acute myocardial edema with high diagnostic accuracy: a comparison to T2-weighted cardiovascular magnetic resonance , 2012, Journal of Cardiovascular Magnetic Resonance.

[51]  S. K. White,et al.  Normal variation of magnetic resonance T1 relaxation times in the human population at 1.5 T using ShMOLLI , 2013, Journal of Cardiovascular Magnetic Resonance.

[52]  Nassir Navab,et al.  Inherent Brain Segmentation Quality Control from Fully ConvNet Monte Carlo Sampling , 2018, MICCAI.

[53]  Ben Glocker,et al.  Automatic Quality Control of Cardiac MRI Segmentation in Large-Scale Population Imaging , 2017, MICCAI.

[54]  Teng-Yi Huang,et al.  Automatic regional analysis of myocardial native T1 values: left ventricle segmentation and AHA parcellations , 2017, The International Journal of Cardiovascular Imaging.

[55]  Balaji Lakshminarayanan,et al.  Deep Ensembles: A Loss Landscape Perspective , 2019, ArXiv.

[56]  Stefan Neubauer,et al.  T(1) mapping for the diagnosis of acute myocarditis using CMR: comparison to T2-weighted and late gadolinium enhanced imaging. , 2013, JACC. Cardiovascular imaging.