Segmentation of carotid arterial walls using neural networks

作     者：Daniel D Samber Sarayu RamachANDran Anoop Sahota Sonum Naidu Alison Pruzan Zahi A Fayad Venkatesh Mani 

作者机构：Translational and Molecular Imaging Institute (TMII) Icahn School of Medicine at Mount SinaiNew York 

出 版 物：《World Journal of Radiology》 (世界放射学杂志（英文版）（电子版）)

年 卷 期：2020年第12卷第1期

页      面：1-9页

学科分类：08[工学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：Supported by American Heart Association Grant in Aid Founders Affiliate No.17GRNT33420119(Mani V) NIH NHLBI 2R01HL070121(Fayad ZA)and NIH NHLBI 1R01HL135878(Fayad ZA) 

主　　题：Carotid arteries Segmentation Convolutional neural network Magnetic resonance imaging Vessel wall 

摘      要：BACKGROUND Automated,accurate,objective,and quantitative medical image segmentation has remained a challenging goal in computer science since its *** study applies the technique of convolutional neural networks(CNNs)to the task of segmenting carotid arteries to aid in the assessment of *** To investigate CNN’s utility as an ancillary tool for researchers who require accurate segmentation of carotid *** An expert reader delineated vessel wall boundaries on 4422 axial T2-weighted magnetic resonance images of bilateral carotid arteries from 189 subjects with clinically evident atherosclerotic disease.A portion of this dataset was used to train two CNNs(one to segment the vessel lumen and the other to segment the vessel wall)with the remaining portion used to test the algorithm’s efficacy by comparing CNN segmented images with those of an expert *** quantitative assessment between automated and manual segmentations was determined by computing the DICE coefficient for each pair of segmented images in the test dataset for each CNN *** average DICE coefficient for the test dataset(CNN segmentations compared to expert’s segmentations)was 0.96 for the lumen and 0.87 for the vessel *** correlation values and the intra-class correlation coefficient(ICC)were computed for the lumen(Pearson=0.98,ICC=0.98)and vessel wall(Pearson=0.88,ICC=0.86)***-Altman plots of area measurements for the CNN and expert readers indicate good agreement with a mean bias of 1%-8%.CONCLUSION Although the technique produces reasonable results that are on par with expert human assessments,our application requires human supervision and monitoring to ensure consistent *** intend to deploy this algorithm as part of a software platform to lessen researchers’workload to more quickly obtain reliable results.