干旱区生态环境知识资源中心(Arid): Generating Longitudinal Atrophy Evaluation Datasets on Brain Magnetic Resonance Images Using Convolutional Neural Networks and Segmentation Priors

Arid

DOI	10.1007/s12021-020-09499-z
	Generating Longitudinal Atrophy Evaluation Datasets on Brain Magnetic Resonance Images Using Convolutional Neural Networks and Segmentation Priors
	Bernal, Jose; Valverde, Sergi; Kushibar, Kaisar; Cabezas, Mariano; Oliver, Arnau; Llado, Xavier
通讯作者	Bernal, J (corresponding author), Univ Girona, Comp Vis & Robot Inst, Girona, Spain.
来源期刊	NEUROINFORMATICS
ISSN	1539-2791
EISSN	1559-0089
出版年	2021
卷号	19 期号:3 页码:477-492
英文摘要	Brain atrophy quantification plays a fundamental role in neuroinformatics since it permits studying brain development and neurological disorders. However, the lack of a ground truth prevents testing the accuracy of longitudinal atrophy quantification methods. We propose a deep learning framework to generate longitudinal datasets by deforming T1-w brain magnetic resonance imaging scans as requested through segmentation maps. Our proposal incorporates a cascaded multi-path U-Net optimised with a multi-objective loss which allows its paths to generate different brain regions accurately. We provided our model with baseline scans and real follow-up segmentation maps from two longitudinal datasets, ADNI and OASIS, and observed that our framework could produce synthetic follow-up scans that matched the real ones (Total scans= 584; Median absolute error: 0.03 +/- 0.02; Structural similarity index: 0.98 +/- 0.02; Dice similarity coefficient: 0.95 +/- 0.02; Percentage of brain volume change: 0.24 +/- 0.16; Jacobian integration: 1.13 +/- 0.05). Compared to two relevant works generating brain lesions using U-Nets and conditional generative adversarial networks (CGAN), our proposal outperformed them significantly in most cases (p < 0.01), except in the delineation of brain edges where the CGAN took the lead (Jacobian integration: Ours - 1.13 +/- 0.05 vs CGAN - 1.00 +/- 0.02; p < 0.01). We examined whether changes induced with our framework were detected by FAST, SPM, SIENA, SIENAX, and the Jacobian integration method. We observed that induced and detected changes were highly correlated (Adj. R-2 > 0.86). Our preliminary results on harmonised datasets showed the potential of our framework to be applied to various data collections without further adjustment.
英文关键词	Cerebral atrophy Longitudinal atrophy synthesis Image generation Convolutional neural networks Brain MRI
类型	Article
语种	英语
收录类别	SCI-E
WOS记录号	WOS:000604154700001
WOS关键词	MULTIPLE-SCLEROSIS ; MATTER ATROPHY ; VOLUME CHANGES ; ROBUST ; ACCURATE ; MS ; QUANTIFICATION ; OPTIMIZATION ; REGISTRATION ; GRAY
WOS类目	Computer Science, Interdisciplinary Applications ; Neurosciences
WOS研究方向	Computer Science ; Neurosciences & Neurology
资源类型	期刊论文
条目标识符	http://119.78.100.177/qdio/handle/2XILL650/351204
作者单位	[Bernal, Jose; Valverde, Sergi; Kushibar, Kaisar; Cabezas, Mariano; Oliver, Arnau; Llado, Xavier] Univ Girona, Comp Vis & Robot Inst, Girona, Spain
推荐引用方式 GB/T 7714	Bernal, Jose,Valverde, Sergi,Kushibar, Kaisar,et al. Generating Longitudinal Atrophy Evaluation Datasets on Brain Magnetic Resonance Images Using Convolutional Neural Networks and Segmentation Priors[J],2021,19(3):477-492.
APA	Bernal, Jose,Valverde, Sergi,Kushibar, Kaisar,Cabezas, Mariano,Oliver, Arnau,&Llado, Xavier.(2021).Generating Longitudinal Atrophy Evaluation Datasets on Brain Magnetic Resonance Images Using Convolutional Neural Networks and Segmentation Priors.NEUROINFORMATICS,19(3),477-492.
MLA	Bernal, Jose,et al."Generating Longitudinal Atrophy Evaluation Datasets on Brain Magnetic Resonance Images Using Convolutional Neural Networks and Segmentation Priors".NEUROINFORMATICS 19.3(2021):477-492.