Julia Karr1, Michael Cohen2, Samuel A McQuiston3, Teja Poorsala4, Christopher Malozzi2. 1. Departments of Mechanical Engineering and Pharmacology, University of South Alabama, Mobile, AL, USA. 2. Department of Cardiology, College of Medicine, University of South Alabama, Mobile, AL, USA. 3. Department of Radiology, University of South Alabama, Mobile, AL, USA. 4. Departments of Oncology and Hematology, University of South Alabama, Mobile, AL, USA.
Abstract
OBJECTIVE: Left-ventricular (LV) strain measurements with the Displacement Encoding with Stimulated Echoes (DENSE) MRI sequence provide accurate estimates of cardiotoxicity damage related to chemotherapy for breast cancer. This study investigated an automated and supervised deep convolutional neural network (DCNN) model for LV chamber quantification before strain analysis in DENSE images. METHODS: The DeepLabV3 +DCNN with three versions of ResNet-50 backbone was designed to conduct chamber quantification on 42 female breast cancer data sets. The convolutional layers in the three ResNet-50 backbones were varied as non-atrous, atrous and modified, atrous with accuracy improvements like using Laplacian of Gaussian filters. Parameters such as LV end-diastolic diameter (LVEDD) and ejection fraction (LVEF) were quantified, and myocardial strains analyzed with the Radial Point Interpolation Method (RPIM). Myocardial classification was validated with the performance metrics of accuracy, Dice, average perpendicular distance (APD) and others. Repeated measures ANOVA and intraclass correlation (ICC) with Cronbach's α (C-Alpha) tests were conducted between the three DCNNs and a vendor tool on chamber quantification and myocardial strain analysis. RESULTS: Validation results in the same test-set for myocardial classification were accuracy = 97%, Dice = 0.92, APD = 1.2 mm with the modified ResNet-50, and accuracy = 95%, Dice = 0.90, APD = 1.7 mm with the atrous ResNet-50. The ICC results between the modified ResNet-50, atrous ResNet-50 and vendor-tool were C-Alpha = 0.97 for LVEF (55±7%, 54±7%, 54±7%, p = 0.6), and C-Alpha = 0.87 for LVEDD (4.6 ± 0.3 cm, 4.6 ± 0.3 cm, 4.6 ± 0.4 cm, p = 0.7). CONCLUSION: Similar performance metrics and equivalent parameters obtained from comparisons between the atrous networks and vendor tool show that segmentation with the modified, atrous DCNN is applicable for automated LV chamber quantification and subsequent strain analysis in cardiotoxicity. ADVANCES IN KNOWLEDGE: A novel deep-learning technique for segmenting DENSE images was developed and validated for LV chamber quantification and strain analysis in cardiotoxicity detection.
OBJECTIVE: Left-ventricular (LV) strain measurements with the Displacement Encoding with Stimulated Echoes (DENSE) MRI sequence provide accurate estimates of cardiotoxicity damage related to chemotherapy for breast cancer. This study investigated an automated and supervised deep convolutional neural network (DCNN) model for LV chamber quantification before strain analysis in DENSE images. METHODS: The DeepLabV3 +DCNN with three versions of ResNet-50 backbone was designed to conduct chamber quantification on 42 female breast cancer data sets. The convolutional layers in the three ResNet-50 backbones were varied as non-atrous, atrous and modified, atrous with accuracy improvements like using Laplacian of Gaussian filters. Parameters such as LV end-diastolic diameter (LVEDD) and ejection fraction (LVEF) were quantified, and myocardial strains analyzed with the Radial Point Interpolation Method (RPIM). Myocardial classification was validated with the performance metrics of accuracy, Dice, average perpendicular distance (APD) and others. Repeated measures ANOVA and intraclass correlation (ICC) with Cronbach's α (C-Alpha) tests were conducted between the three DCNNs and a vendor tool on chamber quantification and myocardial strain analysis. RESULTS: Validation results in the same test-set for myocardial classification were accuracy = 97%, Dice = 0.92, APD = 1.2 mm with the modified ResNet-50, and accuracy = 95%, Dice = 0.90, APD = 1.7 mm with the atrous ResNet-50. The ICC results between the modified ResNet-50, atrous ResNet-50 and vendor-tool were C-Alpha = 0.97 for LVEF (55±7%, 54±7%, 54±7%, p = 0.6), and C-Alpha = 0.87 for LVEDD (4.6 ± 0.3 cm, 4.6 ± 0.3 cm, 4.6 ± 0.4 cm, p = 0.7). CONCLUSION: Similar performance metrics and equivalent parameters obtained from comparisons between the atrous networks and vendor tool show that segmentation with the modified, atrous DCNN is applicable for automated LV chamber quantification and subsequent strain analysis in cardiotoxicity. ADVANCES IN KNOWLEDGE: A novel deep-learning technique for segmenting DENSE images was developed and validated for LV chamber quantification and strain analysis in cardiotoxicity detection.
Authors: Qian Tao; Wenjun Yan; Yuanyuan Wang; Elisabeth H M Paiman; Denis P Shamonin; Pankaj Garg; Sven Plein; Lu Huang; Liming Xia; Marek Sramko; Jarsolav Tintera; Albert de Roos; Hildo J Lamb; Rob J van der Geest Journal: Radiology Date: 2018-10-09 Impact factor: 11.105
Authors: Xiaodong Zhong; Bruce S Spottiswoode; Craig H Meyer; Christopher M Kramer; Frederick H Epstein Journal: Magn Reson Med Date: 2010-10 Impact factor: 4.668
Authors: Mir Henglin; Gillian Stein; Pavel V Hushcha; Jasper Snoek; Alexander B Wiltschko; Susan Cheng Journal: Circ Cardiovasc Imaging Date: 2017-10 Impact factor: 7.792
Authors: Avan Suinesiaputra; Pierre Ablin; Xenia Alba; Martino Alessandrini; Jack Allen; Wenjia Bai; Serkan Cimen; Peter Claes; Brett R Cowan; Jan D'hooge; Nicolas Duchateau; Jan Ehrhardt; Alejandro F Frangi; Ali Gooya; Vicente Grau; Karim Lekadir; Allen Lu; Anirban Mukhopadhyay; Ilkay Oksuz; Nripesh Parajali; Xavier Pennec; Marco Pereanez; Catarina Pinto; Paolo Piras; Marc-Michel Rohe; Daniel Rueckert; Dennis Saring; Maxime Sermesant; Kaleem Siddiqi; Mahdi Tabassian; Luciano Teresi; Sotirios A Tsaftaris; Matthias Wilms; Alistair A Young; Xingyu Zhang; Pau Medrano-Gracia Journal: IEEE J Biomed Health Inform Date: 2017-01-17 Impact factor: 5.772