A fully automated pipeline for brain structure segmentation in multiple sclerosis

被引:5
|
作者
Gonzalez-Villa, Sandra [1 ,2 ]
Oliver, Arnau [1 ]
Huo, Yuankai [2 ]
Llado, Xavier [1 ]
Landman, Bennett A. [2 ]
机构
[1] Univ Girona, Inst Comp Vis & Robot, P-4,Campus Montilivi, Girona 17003, Spain
[2] Vanderbilt Univ, Elect Engn, 221 Kirkland Hall, Nashville, TN 37235 USA
关键词
Brain structures; Parcellation; Multiple sclerosis lesions; Segmentation; MRI; Multi-atlas; Label fusion; ATROPHY; CLASSIFICATION; LESIONS; FUSION; VOLUME;
D O I
10.1016/j.nicl.2020.102306
中图分类号
R445 [影像诊断学];
学科分类号
100207 ;
摘要
Accurate volume measurements of the brain structures are important for treatment evaluation and disease follow-up in multiple sclerosis (MS) patients. With the aim of obtaining reproducible measurements and avoiding the intra-/inter-rater variability that manual delineations introduce, several automated brain structure segmentation strategies have been proposed in recent years. However, most of these strategies tend to be affected by the abnormal MS lesion intensities, which corrupt the structure segmentation result. To address this problem, we recently reformulated two label fusion strategies of the state of the art, improving their segmentation performance on the lesion areas. Here, we integrate these reformulated strategies in a completely automated pipeline that includes pre-processing (inhomogeneity correction and intensity normalization), atlas selection, masked registration and label fusion, and combine them with an automated lesion segmentation method of the state of the art. We study the effect of automating the lesion mask acquisition on the structure segmentation result, analyzing the output of the proposed pipeline when used in combination with manually and automatically segmented lesion masks. We further analyze the effect of those masks on the segmentation result of the original label fusion strategies when combined with the well-established pre-processing step of lesion filling. The experiments performed show that, when the original methods are used to segment the lesion-filled images, significant structure volume differences are observed in a comparison between manually and automatically segmented lesion masks. The results indicate a mean volume decrease of 1.13% +/- 1.93 in the cerebrospinal fluid, and a mean volume increase of 0.13% +/- 0.14 and 0.05% +/- 0.08 in the cerebral white matter and cerebellar gray matter, respectively. On the other hand, no significant volume differences were found when the proposed automated pipeline was used for segmentation, which demonstrates its robustness against variations in the lesion mask used.
引用
收藏
页数:8
相关论文
共 50 条
  • [21] BrainSec: Automated Brain Tissue Segmentation Pipeline for Scalable Neuropathological Analysis
    Lai, Zhengfeng
    Oliveira, Luca Cerny
    Guo, Runlin
    Xu, Wenda
    Hu, Zin
    Mifflin, Kelsey
    Decarli, Charles
    Cheung, Sen-Ching
    Chuah, Chen-Nee
    Dugger, Brittany N.
    IEEE ACCESS, 2022, 10 : 49064 - 49079
  • [22] A fully automated deep learning pipeline for high-throughput colony segmentation and classification
    Carl, Sarah H.
    Duempelmann, Lea
    Shimada, Yukiko
    Buhler, Marc
    BIOLOGY OPEN, 2020, 9 (06):
  • [23] Fully-automated deep learning pipeline for segmentation and classification of breast ultrasound images
    Podda, Alessandro Sebastian
    Balia, Riccardo
    Barra, Silvio
    Carta, Salvatore
    Fenu, Gianni
    Piano, Leonardo
    JOURNAL OF COMPUTATIONAL SCIENCE, 2022, 63
  • [24] Automated segmentation of multiple sclerosis lesion subtypes with multichannel MRI
    Wu, Ying
    Warfield, Simon K.
    Tan, I. Leng
    Wells, William M., III
    Meier, Dominik S.
    van Schijndel, Ronald A.
    Barkhof, Frederik
    Guttmann, Charles R. G.
    NEUROIMAGE, 2006, 32 (03) : 1205 - 1215
  • [25] A CNN Based Algorithm for the Automated Segmentation of Multiple Sclerosis Lesions
    Bilotta, Eleonora
    Cerasa, Antonio
    Pantano, Pietro
    Quattrone, Aldo
    Staino, Andrea
    Stramandinoli, Francesca
    APPLICATIONS OF EVOLUTIONARY COMPUTATION, PT I, PROCEEDINGS, 2010, 6024 : 211 - +
  • [26] Automated Segmentation Of Multiple Sclerosis Lesions Using Statistical Approach
    Sharma, Yamini
    Meghrajani, Yogesh K.
    2015 INTERNATIONAL CONFERENCE ON INNOVATIONS IN INFORMATION, EMBEDDED AND COMMUNICATION SYSTEMS (ICIIECS), 2015,
  • [27] Automated segmentation of multiple sclerosis lesions by model outlier detection
    Van Leemput, K
    Maes, F
    Vandermeulen, D
    Colchester, A
    Suetens, P
    IEEE TRANSACTIONS ON MEDICAL IMAGING, 2001, 20 (08) : 677 - 688
  • [28] Automated Segmentation of Multiple Sclerosis Brain Lesions - a Simple Step in the Complicated Problem of Clinical Imaging Standardization
    Yeung, Stephen
    Leyden, Kelly
    Luo, Weidong
    Kinkel, Revere
    MULTIPLE SCLEROSIS JOURNAL, 2018, 24 : 74 - 74
  • [29] Whole-brain atrophy in multiple sclerosis measured by automated versus semiautomated MR imaging segmentation
    Sharma, J
    Sanfilipo, MP
    Benedict, RHB
    Weinstock-Guttman, B
    Munschauer, FE
    Bakshi, R
    AMERICAN JOURNAL OF NEURORADIOLOGY, 2004, 25 (06) : 985 - 996
  • [30] Impact and extent of hypointense lesion volume misclassification on fully automated grey and white matter segmentation in patients with multiple sclerosis
    Dwyer, MG
    Srinivasaraghavan, B
    Watts, KL
    Abdelrahman, N
    Hussein, S
    Zivadinov, R
    MULTIPLE SCLEROSIS, 2005, 11 : S149 - S149