Magnus Ziegler1,2, Jesper Alfraeus3, Mariana Bustamante3,4, Elin Good3,4,5, Jan Engvall3,4,6, Ebo de Muinck3,4,5, Petter Dyverfeldt3,4. 1. Division of Cardiovascular Medicine, Department of Health, Medicine and Caring Sciences, Linköping University, Linköping, Sweden. magnus.ziegler@liu.se. 2. Center for Medical Image Science and Visualization (CMIV), Linköping University, Linköping, Sweden. magnus.ziegler@liu.se. 3. Division of Cardiovascular Medicine, Department of Health, Medicine and Caring Sciences, Linköping University, Linköping, Sweden. 4. Center for Medical Image Science and Visualization (CMIV), Linköping University, Linköping, Sweden. 5. Department of Cardiology, Department of Health, Medicine and Caring Sciences, Linköping University, Linköping, Sweden. 6. Department of Clinical Physiology, Department of Health, Medicine and Caring Sciences, Linköping University, Linköping, Sweden.
Abstract
BACKGROUND: Non-invasive imaging is of interest for tracking the progression of atherosclerosis in the carotid bifurcation, and segmenting this region into its constituent branch arteries is necessary for analyses. The purpose of this study was to validate and demonstrate a method for segmenting the carotid bifurcation into the common, internal, and external carotid arteries (CCA, ICA, ECA) in contrast-enhanced MR angiography (CE-MRA) data. METHODS: A segmentation pipeline utilizing a convolutional neural network (DeepMedic) was tailored and trained for multi-class segmentation of the carotid arteries in CE-MRA data from the Swedish CardioPulmonsary bioImage Study (SCAPIS). Segmentation quality was quantitatively assessed using the Dice similarity coefficient (DSC), Matthews Correlation Coefficient (MCC), F2, F0.5, and True Positive Ratio (TPR). Segmentations were also assessed qualitatively, by three observers using visual inspection. Finally, geometric descriptions of the carotid bifurcations were generated for each subject to demonstrate the utility of the proposed segmentation method. RESULTS: Branch-level segmentations scored DSC = 0.80 ± 0.13, MCC = 0.80 ± 0.12, F2 = 0.82 ± 0.14, F0.5 = 0.78 ± 0.13, and TPR = 0.84 ± 0.16, on average in a testing cohort of 46 carotid bifurcations. Qualitatively, 61% of segmentations were judged to be usable for analyses without adjustments in a cohort of 336 carotid bifurcations without ground-truth. Carotid artery geometry showed wide variation within the whole cohort, with CCA diameter 8.6 ± 1.1 mm, ICA 7.5 ± 1.4 mm, ECA 5.7 ± 1.0 mm and bifurcation angle 41 ± 21°. CONCLUSION: The proposed segmentation method automatically generates branch-level segmentations of the carotid arteries that are suitable for use in further analyses and help enable large-cohort investigations.
BACKGROUND: Non-invasive imaging is of interest for tracking the progression of atherosclerosis in the carotid bifurcation, and segmenting this region into its constituent branch arteries is necessary for analyses. The purpose of this study was to validate and demonstrate a method for segmenting the carotid bifurcation into the common, internal, and external carotid arteries (CCA, ICA, ECA) in contrast-enhanced MR angiography (CE-MRA) data. METHODS: A segmentation pipeline utilizing a convolutional neural network (DeepMedic) was tailored and trained for multi-class segmentation of the carotid arteries in CE-MRA data from the Swedish CardioPulmonsary bioImage Study (SCAPIS). Segmentation quality was quantitatively assessed using the Dice similarity coefficient (DSC), Matthews Correlation Coefficient (MCC), F2, F0.5, and True Positive Ratio (TPR). Segmentations were also assessed qualitatively, by three observers using visual inspection. Finally, geometric descriptions of the carotid bifurcations were generated for each subject to demonstrate the utility of the proposed segmentation method. RESULTS: Branch-level segmentations scored DSC = 0.80 ± 0.13, MCC = 0.80 ± 0.12, F2 = 0.82 ± 0.14, F0.5 = 0.78 ± 0.13, and TPR = 0.84 ± 0.16, on average in a testing cohort of 46 carotid bifurcations. Qualitatively, 61% of segmentations were judged to be usable for analyses without adjustments in a cohort of 336 carotid bifurcations without ground-truth. Carotid artery geometry showed wide variation within the whole cohort, with CCA diameter 8.6 ± 1.1 mm, ICA 7.5 ± 1.4 mm, ECA 5.7 ± 1.0 mm and bifurcation angle 41 ± 21°. CONCLUSION: The proposed segmentation method automatically generates branch-level segmentations of the carotid arteries that are suitable for use in further analyses and help enable large-cohort investigations.
Entities:
Keywords:
Atherosclerosis; Carotid arteries; Contrast-enhanced; Deep learning; Magnetic resonance imaging; Segmentation
Authors: Avan Suinesiaputra; Patrick J H de Koning; Elena Zudilova-Seinstra; Johan H C Reiber; Rob J van der Geest Journal: Int J Cardiovasc Imaging Date: 2011-12-09 Impact factor: 2.357
Authors: Jonathan B Thomas; Luca Antiga; Susan L Che; Jaques S Milner; Dolores A Hangan Steinman; J David Spence; Brian K Rutt; David A Steinman Journal: Stroke Date: 2005-10-13 Impact factor: 7.914
Authors: Hui Tang; Theo van Walsum; Robbert S van Onkelen; Reinhard Hameeteman; Stefan Klein; Michiel Schaap; Fufa L Tori; Quirijn J A van den Bouwhuijsen; Jacqueline C M Witteman; Aad van der Lugt; Lucas J van Vliet; Wiro J Niessen Journal: Med Image Anal Date: 2012-06-19 Impact factor: 8.545
Authors: Willem E Hellings; Wouter Peeters; Frans L Moll; Sebastiaan R D Piers; Jessica van Setten; Peter J Van der Spek; Jean-Paul P M de Vries; Kees A Seldenrijk; Peter C De Bruin; Aryan Vink; Evelyn Velema; Dominique P V de Kleijn; Gerard Pasterkamp Journal: Circulation Date: 2010-04-19 Impact factor: 29.690
Authors: Sandeep Koppal; Marcel Warntjes; Jeremy Swann; Petter Dyverfeldt; Johan Kihlberg; Rodrigo Moreno; Derek Magee; Nicholas Roberts; Helene Zachrisson; Claes Forssell; Toste Länne; Darren Treanor; Ebo D de Muinck Journal: Magn Reson Med Date: 2016-08-11 Impact factor: 4.668
Authors: Hunter R Underhill; Thomas S Hatsukami; Zahi A Fayad; Valentin Fuster; Chun Yuan Journal: Nat Rev Cardiol Date: 2010-01-26 Impact factor: 32.419
Authors: Li Dong; William S Kerwin; Marina S Ferguson; Rui Li; Jinnan Wang; Huijun Chen; Gador Canton; Thomas S Hatsukami; Chun Yuan Journal: J Cardiovasc Magn Reson Date: 2009-12-15 Impact factor: 5.364