Semi-Supervised Intracranial Aneurysm Segmentation from CTA Images via Weight-Perceptual Self-Ensembling Model
作者机构：Shenzhen Institute of Advanced TechnologyChinese Academy of SciencesShenzhen 518055China University of Chinese Academy of SciencesBeijing 100049China Shantou UniversityShantou 515063China Shenzhen Second People′s HospitalShenzhen 518035China Department of Computer Science and EngineeringThe Chinese University of Hong KongHong Kong 999077China
出 版 物：《Journal of Computer Science & Technology》 (计算机科学技术学报（英文版）)
年 卷 期：2023年第38卷第3期
页 面：674-685页
核心收录：
学科分类：08[工学] 080203[工学-机械设计及理论] 0802[工学-机械工程]
基 金：supported by Shenzhen Fundamental Research Program of China under Grant Nos.JCYJ20200109110420626 and JCYJ20200109110208764 the National Natural Science Foundation of China under Grant Nos.U1813204 and 61802385 the Natural Science Foundation of Guangdong of China under Grant No.2021A1515012604 the Clinical Research Project of Shenzhen Municiple Health Commission under Grant No.SZLY2017011
主 题：intracranial aneurysm(IA)segmentation sample weight perception self-ensembling model semi-supervised learning
摘 要：Segmentation of intracranial aneurysm(IA)from computed tomography angiography(CTA)images is of significant importance for quantitative assessment of IA and further surgical *** segmentation of IA is a labor-intensive,time-consuming job and suffers from inter-and intra-observer *** deep neural networks usually requires a large amount of labeled data,while annotating data is very time-consuming for the IA segmentation *** paper presents a novel weight-perceptual self-ensembling model for semi-supervised IA segmentation,which employs unlabeled data by encouraging the predictions of given perturbed input samples to be *** that the quality of consistency targets is not comparable to each other,we introduce a novel sample weight perception module to quantify the quality of different consistency *** proposed module can be used to evaluate the contributions of unlabeled samples during training to force the network to focus on those well-predicted *** have conducted both horizontal and vertical comparisons on the clinical intracranial aneurysm CTA image *** results show that our proposed method can improve at least 3%Dice coefficient over the fully-supervised baseline,and at least 1.7%over other state-of-the-art semi-supervised methods.
维普期刊数据库