Share：
Share this content in WeChat
X
[Chinese][PDF]103689
• Clinical Article •
Quantitative evaluation of the cortical development on neonates based on segmentation of 3D T1WI images using transfer learning
LI Xianjun  CHEN Jian  XIA Jing  WANG Miaomiao  LI Mengxuan  WANG Li  LI Gang  SHEN Dinggang  YANG Jian 

Cite this article as: Li XJ, Chen J, Xia J, et al. Quantitative evaluation of the cortical development on neonates based on segmentation of 3D T1WI images using transfer learning. Chin J Magn Reson Imaging, 2019, 10(10): 736-742. DOI:10.12015/issn.1674-8034.2019.10.004.


[Abstract] Objective: To implement a machine learning-based segmentation method on neonatal T1WI images and assess the cortical structural maturation of preterm and term neonates.Materials and Methods: This work enrolled 50 subjects without any abnormalities on magnetic resonance imaging from the First Affiliated Hospital of Xi'an Jiaotong University, including preterm neonates, preterm newborns at the term equivalent age, and term neonates. The preliminary training of a densely convolution network was performed by using the similar neonatal dataset from the shared database. Segmentation of the local dataset was performed by using this preliminary model. The segmentation results were modified manually by experts. Then the segmentation model was trained again by using the local data from randomly selected 25 subjects. According to the segmentation quality of the validation dataset (sample size: 10), the model parameters were adjusted. Finally, this segmentation model was assessed by using Dice ratios on the testing dataset (sample size: 15). The surface area, cortical thickness, and cortical volume of left and right hemispheres of the brain were extracted based on the cortical reconstruction. Correlations between these metrics and the postmenstrual age were performed by using the Spearman partial correlation. Inter-group differences were evaluated by using the Mann-Whitney U test.Results: The proposed model could effectively segment the gray matter, white matter, and cerebrospinal fluid regions in T1WI images on neonatal brains. This method was feasible on preterm neonates, preterm newborns at the term equivalent age, and term neonates. The dice ratios ranged from 0.93 to 0.99. Significant positive correlations between surface area, cortical volume and the postmenstrual age were observed on both hemispheres (P<0.05). Postmenstrual age-related change patterns of the cortical thickness on left and right hemispheres were different. Except that no significant inter-group difference could be found in cortical thickness, preterm neonates held smaller surface area and cortical volume on both hemispheres than term neonates (P<0.001).Conclusions: It is feasible to implement segmentation of T1WI images on neonatal brains based on a densely connected convolution network. Metrics extracted from the image segmentation and cortical reconstruction could be used to quantitatively assess the cortical structural maturation of neonates. Cortical maturation was delayed in preterm neonates than term neonates on both hemispheres.
[Keywords] magnetic resonance imaging；neonate；cortical development；image segmentation；machine learning

LI Xianjun# Department of Radiology, the First Affiliated Hospital of Xi'an Jiaotong University, Xi'an 710061, China

CHEN Jian# Department of Radiology, University of North Carolina at Chapel Hill, North Carolina, NC 27599-7513, USA; School of Information Science and Engineering, Fujian University of Technology, Fuzhou 350118, China

XIA Jing Department of Radiology, University of North Carolina at Chapel Hill, North Carolina, NC 27599-7513, USA

WANG Miaomiao Department of Radiology, the First Affiliated Hospital of Xi'an Jiaotong University, Xi'an 710061, China

LI Mengxuan Department of Radiology, the First Affiliated Hospital of Xi'an Jiaotong University, Xi'an 710061, China

WANG Li Department of Radiology, University of North Carolina at Chapel Hill, North Carolina, NC 27599-7513, USA

LI Gang Department of Radiology, University of North Carolina at Chapel Hill, North Carolina, NC 27599-7513, USA

SHEN Dinggang* Department of Radiology, University of North Carolina at Chapel Hill, North Carolina, NC 27599-7513, USA

YANG Jian* Department of Radiology, the First Affiliated Hospital of Xi'an Jiaotong University, Xi'an 710061, China

#: These authors contributed equally to this work.

*Correspondence to: Yang J, E-mail: yj1118@mail.xjtu.edu.cn Shen DG, E-mail: dinggang_shen@med.unc.edu

Conflicts of interest   None.

ACKNOWLEDGMENTS  This work was part of National Key Research and Development Program of China No.2016YFC0100300 National Natural Science Foundation of China No.81771810, 81471631, 81171317 Fundamental Research Funds for the Central Universities No.xjj2018265 Fundamental Research Funds of the First Affiliated Hospital of Xi'an Jiaotong University No.2017QN-09
Received  2019-04-30
DOI: 10.12015/issn.1674-8034.2019.10.004
Cite this article as: Li XJ, Chen J, Xia J, et al. Quantitative evaluation of the cortical development on neonates based on segmentation of 3D T1WI images using transfer learning. Chin J Magn Reson Imaging, 2019, 10(10): 736-742. DOI:10.12015/issn.1674-8034.2019.10.004.

[1]
Jha SC, Xia K, Schmitt JE, et al. Genetic influences on neonatal cortical thickness and surface area. Hum Brain Mapp, 2018, 39(12): 4998-5013.
[2]
Lefèvre J, Germanaud D, Dubois J, et al. Are developmental trajectories of cortical folding comparable between cross-sectional datasets of fetuses and preterm newborns?. Cereb Cortex, 2015, 26(7): 99-100.
[3]
Tortora D, Panara V, Mattei PA, et al. Comparing 3 T T1-weighted sequences in identifying hyperintense punctate lesions in preterm neonates. AJNR Am J Neuroradiol, 2015, 36(3): 581-586.
[4]
孙亲利，张育苗，高洁，等．磁共振3D-T1WI序列在新生儿局灶性脑白质损伤病灶检出中的应用．磁共振成像, 2018, 9(11): 11-16.
[5]
Li G, Wang L, Yap PT, et al. Computational neuroanatomy of baby brains: A review. Neuroimage, 2019, 185(1): 906-925.
[6]
Wang L, Gao Y, Shi F, et al. LINKS: Learning-based multi-source integration framework for segmentation of infant brain images. Neuroimage, 2015, 108(1): 160-172.
[7]
Wang L, Shi F, Yap PT, et al. Longitudinally guided level sets for consistent tissue segmentation of neonates. Hum Brain Mapp, 2013, 34(4): 956-972.
[8]
Xianjun L, Jian Y, Jie G, et al. A robust post-processing workflow for datasets with motion artifacts in diffusion kurtosis imaging. PLoS One, 2014, 9(4): e94592.
[9]
Bui TD, Shin J, Moon T. 3D densely convolution networks for volumetric segmentation. Arxiv, 2017, 1709(3): 1-7.
[10]
Wang L, Li G, Shi F, et al. Volume-based analysis of 6-month-old infant brain MRI for Autism biomarker identification and early diagnosis. Med Image Comput Comput Assist Interv, 2018, 11072(1): 411-419.
[11]
Coté CJ, Stephen W. Guidelines for monitoring and management of pediatric patients during and after sedation for diagnostic and therapeutic procedures: an update. Pediatrics, 2006, 118(6): 2587-2602.
[12]
Wang L, Li G, Adeli E, et al. Anatomy-guided joint tissue segmentation and topological correction for 6-month infant brain MRI with risk of autism. Hum Brain Mapp, 2018, 39(6): 2609-2623.
[13]
Cheng B, Liu M, Shen D, et al. Multi-domain transfer learning for early diagnosis of Alzheimer' s disease. Neuroinformatics, 2017, 15(2): 1-18.
[14]
Li G, Wang L, Shi F, et al. Construction of 4D high-definition cortical surface atlases of infants: Methods and applications. Med Image Anal, 2015, 25(1): 22-36.
[15]
Maria KM, Paul A, Latha S, et al. A dynamic 4D probabilistic atlas of the developing brain. Neuroimage, 2011, 54(4): 2750-2763.
[16]
Gilmore JH, Knickmeyer RC, Gao W. Imaging structural and functional brain development in early childhood. Nat Rev Neurosci, 2018, 19(3): 123-137.
[17]
Charles R, Tahani A, Neda R, et al. The premature brain: developmental and lesional anatomy. Neuroradiology, 2013, 55(2): 23-40.
[18]
Paredes MF, James D, Gil-Perotin S, et al. Extensive migration of young neurons into the infant human frontal lobe. Science, 2016, 354(6308): 7073.
[19]
Ghislaine DL, Stanislas D, Lucie HP. Functional neuroimaging of speech perception in infants. Science, 2002, 298(5600): 2013-2015.
[20]
王宏，许建铭．多模态磁共振功能成像在早产儿脑损伤中的应用．磁共振成像, 2016, 7(12): 951-956.
[21]
Zambrana IM, Vollrath ME, Sengpiel V, et al. Preterm delivery and risk for early language delays: a sibling-control cohort study. Int J Epidemiol, 2015, 45(1): 151-159.
[22]
Nam KW, Castellanos N, Simmons A, et al. Alterations in cortical thickness development in preterm-born individuals: Implications for high-order cognitive functions. Neuroimage, 2015, 115(1): 64-75.
[23]
Zoltan N, Hugo L, Chloe H. Effects of preterm birth on cortical thickness measured in adolescence. Cereb Cortex, 2011, 21(2): 300-306.
[24]
Kim H, Lepage C, Maheshwary R, et al. NEOCIVET: Towards accurate morphometry of neonatal gyrification and clinical applications in preterm newborns. Neuroimage, 2016, 138(1): 28-42.
[25]
Farah MJ. The neuroscience of socioeconomic status: Correlates, causes, and consequences. Neuron, 2017, 96(1): 56-71.
[26]
Ashburner J, Friston KJ. Unified segmentation. Neuroimage, 2005, 26(3): 839-851.
[27]
Shi F, Yap PT, Wu G, et al. Infant brain atlases from neonates to 1- and 2-year-olds. PLoS One, 2011, 6(4): e18746.

PREV Analysis of MRI configuration and operation status of county hospitals in Yunnan province
NEXT Study of the relationship between vascular wall changes of vertebrobasilar atherosclerosis and ischemic stroke by HRMR-VWI
  


LINK


Tel & Fax: +8610-67113815    E-mail: editor@cjmri.cn