CIVET-Macaque: an automated pipeline for MRI-based cortical surface generation and cortical thickness in macaques

Lepage, Claude; Wagstyl, Konrad; Jung, Benjamin; Seidlitz, Jakob; Sponheim, Caleb; Ungerleider, Leslie G.; Wang, Xindi; Evans, Alan C.; Messinger, Adam

doi:10.1101/2020.08.04.237149

Cited by 3 publications

(5 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As the recent explosion of MRI data sharing in Nonhuman Primate (NHP) scales the amounts and diversity of data available for NHP imaging studies, researchers are having to overcome key challenges in preprocessing, which will otherwise slow the pace of progress (Autio et al, 2020a; Lepage et al, 2020; Messinger et al, 2020; Milham et al, 2018). Among them is one of the fundamental preprocessing steps - brain extraction (also referred to as skull-stripping) (Seidlitz et al, 2018; Tasserie et al, 2020; Zhao et al, 2018).…”

Section: Introductionmentioning

confidence: 99%

“…In both human and NHP MRI pipelines, brain extraction is often among the first early preprocessing steps (Esteban et al, 2019; Glasser et al, 2013; Seidlitz et al, 2018; Tasserie et al, 2020; Xu et al, 2015). By removing the non-brain tissue, brain extraction dramatically improves the accuracy of later steps, such as anatomy-based brain registration, pial surface reconstruction, and cross-modality coregistration (e.g., functional MRI, diffusion MRI) (Acosta-Cabronero et al, 2008; Autio et al, 2020a; Lepage et al, 2020; Seidlitz et al, 2018). In humans, automated brain extraction tools have been developed (e.g., the Brain Extraction Tool [BET] in FSL, 3dSkullStrip in AFNI, the Hybrid Watershed Algorithm [HWA] in FreeSurfer, etc.)…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

U-Net Model for Brain Extraction: Trained on Humans for Transfer to Non-human Primates

Wang

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Brain extraction (a.k.a. skull stripping) is a fundamental step in the neuroimaging pipeline as it can affect the accuracy of downstream preprocess such as image registration, tissue classification, etc. Most brain extraction tools have been mainly orientated for human data and are often challenging for non-human primates (NHP). In recent attempts to improve the performance in NHP, deep learning models appear to outperform the traditional tools. However, given the minimal sample size of most NHP studies and notable variations in data quality, the deep learning models are very rarely applied in multi-site samples in NHP imaging. To overcome this challenge, we propose to use transfer-learning framework that leverages a large human imaging dataset to pretrain a convolutional neural network (i.e. U-Net Model), and then transferred to NHP data using a small NHP training sample. The resulting transfer-learning model converged faster and achieved more accurate performance than a similar U-Net Model trained exclusively on NHP samples. We improved the generalizability of the model by upgrading the transfer-learned model using additional training datasets from multiple research sites in the Primate Data-Exchange (PRIME-DE) consortium. Our final model outperformed brain extraction routines from popular MRI packages (AFNI, FSL, and FreeSurfer) across multiple heterogeneous multiple sites from PRIME-DE with less computational cost (20s~10min). Our model, code, and the skull-stripped mask repository of 136 macaque monkeys are publicly available for unrestricted use by the neuroimaging community at https://github.com/HumanBrainED/NHP-BrainExtraction.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

U-Net Model for Brain Extraction: Trained on Humans for Transfer to Non-human Primates

Wang

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…We use two popular methods of the fully automated cortical thickness estimation, CIVET ( MacDonald et al, 2000 ; Zijdenbos et al, 2002 ; Tohka et al, 2004 ; Kim et al, 2005 ; Lerch et al, 2005 ; Lee J. et al, 2006 ; Lee J. K. et al, 2006 ; Lepage et al, 2020 ) and FreeSurfer ( Dale and Sereno, 1993 ; Dale et al, 1999 ; Fischl et al,1999a,b , 2001 , 2002 ; Fischl and Dale, 2000 ; Fischl, 2012 ), to illustrate the application of our proposed platform in comparing either different pipelines or different versions of the same pipeline. Key among our objectives is to evaluate the ability of each pipeline to detect the artificial lesions described above by measuring specificity and sensitivity of detection.…”

Section: Methodsmentioning

confidence: 99%

A Simulation Toolkit for Testing the Sensitivity and Accuracy of Corticometry Pipelines

Omidyeganeh

Khalili-Mahani

Bermudez

et al. 2021

Front. Neuroinform.

Self Cite

View full text Add to dashboard Cite

In recent years, the replicability of neuroimaging findings has become an important concern to the research community. Neuroimaging pipelines consist of myriad numerical procedures, which can have a cumulative effect on the accuracy of findings. To address this problem, we propose a method for simulating artificial lesions in the brain in order to estimate the sensitivity and specificity of lesion detection, using different automated corticometry pipelines. We have applied this method to different versions of two widely used neuroimaging pipelines (CIVET and FreeSurfer), in terms of coefficients of variation; sensitivity and specificity of detecting lesions in 4 different regions of interest in the cortex, while introducing variations to the lesion size, the blurring kernel used prior to statistical analyses, and different thickness metrics (in CIVET). These variations are tested in a between-subject design (in two random groups, with and without lesions, using T1-weigted MRIs of 152 individuals from the International Consortium of Brain Mapping (ICBM) dataset) and in a within-subject pre-/post-lesion design [using 21 T1-Weighted MRIs of a single adult individual, scanned in the Infant Brain Imaging Study (IBIS)]. The simulation method is sensitive to partial volume effect and lesion size. Comparisons between pipelines illustrate the ability of this method to uncover differences in sensitivity and specificity of lesion detection. We propose that this method be adopted in the workflow of software development and release.

show abstract

“…Brain extraction is one of the initial steps of MRI image processing (Esteban et al, 2019;Tasserie et al, 2020). By removing non-brain tissues (skull, muscle, eye, dura mater, external blood vessels, and nerves), the accuracy of brain image processing steps can be improved, such as anatomy-based brain registration, meningeal surface reconstruction, brain volume measurement, and tissue recognition (Xi et al, 2019a,b;Autio et al, 2020;Lepage et al, 2021). However, the performance of existing brain extraction tools is lacking when applied to the macaque brain (Zhao et al, 2018).…”

Section: Introductionmentioning

confidence: 99%

“…(3) Different macaque data collection sites use specific collection protocols and equipment (Milham et al, 2018), resulting in significant differences in the data quality and characteristics. To solve these challenges, other researchers have proposed methods for non-human primate data [i.e., a new option "-monkey" in AFNI (Cox, 1996), registration methods (Lohmeier et al, 2019;Jung et al, 2021)]. However, the final results mostly require manual intervention.…”

Section: Introductionmentioning

confidence: 99%

HC-Net: A hybrid convolutional network for non-human primate brain extraction

Hong

Wang²,

Shang³

et al. 2023

Front. Comput. Neurosci.

View full text Add to dashboard Cite

Brain extraction (skull stripping) is an essential step in the magnetic resonance imaging (MRI) analysis of brain sciences. However, most of the current brain extraction methods that achieve satisfactory results for human brains are often challenged by non-human primate brains. Due to the small sample characteristics and the nature of thick-slice scanning of macaque MRI data, traditional deep convolutional neural networks (DCNNs) are unable to obtain excellent results. To overcome this challenge, this study proposed a symmetrical end-to-end trainable hybrid convolutional neural network (HC-Net). It makes full use of the spatial information between adjacent slices of the MRI image sequence and combines three consecutive slices from three axes for 3D convolutions, which reduces the calculation consumption and promotes accuracy. The HC-Net consists of encoding and decoding structures of 3D convolutions and 2D convolutions in series. The effective use of 2D convolutions and 3D convolutions relieves the underfitting of 2D convolutions to spatial features and the overfitting of 3D convolutions to small samples. After evaluating macaque brain data from different sites, the results showed that HC-Net performed better in inference time (approximately 13 s per volume) and accuracy (mean Dice coefficient reached 95.46%). The HC-Net model also had good generalization ability and stability in different modes of brain extraction tasks.

show abstract

CIVET-Macaque: an automated pipeline for MRI-based cortical surface generation and cortical thickness in macaques

Cited by 3 publications

References 48 publications

U-Net Model for Brain Extraction: Trained on Humans for Transfer to Non-human Primates

U-Net Model for Brain Extraction: Trained on Humans for Transfer to Non-human Primates

A Simulation Toolkit for Testing the Sensitivity and Accuracy of Corticometry Pipelines

HC-Net: A hybrid convolutional network for non-human primate brain extraction

Contact Info

Product

Resources

About