Multiple Atlases Research Articles

Browsing Multiple Subjects When the Atlas Adaptation Cannot Be Achieved via a Warping Strategy.

Brain mapping studies often need to identify brain structures or functional circuits into a set of individual brains. To this end, multiple atlases have been published to represent such structures based on different modalities, subject sets, and techniques. The mainstream approach to exploit these atlases consists in spatially deforming each individual data onto a given atlas using dense deformation fields, which supposes the existence of a continuous mapping between atlases and individuals. However, this continuity is not always verified, and this “iconic” approach has limits. We present in this study an alternative, complementary, “structural” approach, which consists in extracting structures from the individual data, and comparing them without deformation. A “structural atlas” is thus a collection of annotated individual data with a common structure nomenclature. It may be used to characterize structure shape variability across individuals or species, or to train machine learning systems. This study exhibits Anatomist, a powerful structural 3D visualization software dedicated to building, exploring, and editing structural atlases involving a large number of subjects. It has been developed primarily to decipher the cortical folding variability; cortical sulci vary enormously in both size and shape, and some may be missing or have various topologies, which makes iconic approaches inefficient to study them. We, therefore, had to build structural atlases for cortical sulci, and use them to train sulci identification algorithms. Anatomist can display multiple subject data in multiple views, supports all kinds of neuroimaging data, including compound structural object graphs, handles arbitrary coordinate transformation chains between data, and has multiple display features. It is designed as a programming library in both C++ and Python languages, and may be extended or used to build dedicated custom applications. Its generic design makes all the display and structural aspects used to explore the variability of the cortical folding pattern work in other applications, for instance, to browse axonal fiber bundles, deep nuclei, functional activations, or other kinds of cortical parcellations. Multimodal, multi-individual, or inter-species display is supported, and adaptations to large scale screen walls have been developed. These very original features make it a unique viewer for structural atlas browsing.

Multi-Scale Graph Representation Learning for Autism Identification With Functional MRI.

Resting-state functional MRI (rs-fMRI) has been widely used for the early diagnosis of autism spectrum disorder (ASD). With rs-fMRI, the functional connectivity networks (FCNs) are usually constructed for representing each subject, with each element representing the pairwise relationship between brain region-of-interests (ROIs). Previous studies often first extract handcrafted network features (such as node degree and clustering coefficient) from FCNs and then construct a prediction model for ASD diagnosis, which largely requires expert knowledge. Graph convolutional networks (GCNs) have recently been employed to jointly perform FCNs feature extraction and ASD identification in a data-driven manner. However, existing studies tend to focus on the single-scale topology of FCNs by using one single atlas for ROI partition, thus ignoring potential complementary topology information of FCNs at different spatial scales. In this paper, we develop a multi-scale graph representation learning (MGRL) framework for rs-fMRI based ASD diagnosis. The MGRL consists of three major components: (1) multi-scale FCNs construction using multiple brain atlases for ROI partition, (2) FCNs representation learning via multi-scale GCNs, and (3) multi-scale feature fusion and classification for ASD diagnosis. The proposed MGRL is evaluated on 184 subjects from the public Autism Brain Imaging Data Exchange (ABIDE) database with rs-fMRI scans. Experimental results suggest the efficacy of our MGRL in FCN feature extraction and ASD identification, compared with several state-of-the-art methods.

Comparison of atlas-based auto-segmentation accuracy for radiotherapy in prostate cancer.

Atlas-based auto-segmentation (ABS) procedure used in radiotherapy can be classified into two groups, one using one atlas per patient (sSM) and the other using multiple atlases (sMM). This study evaluated auto-contouring accuracy and contouring time in patients with prostate cancer using the two procedures. The Dice similarity coefficient of sMM was significantly better than that of sSM (prostate [median, 0.81 (range, 0.66-0.91) vs. 0.64 (0.27-0.71), p<0.01], seminal vesicles [0.49 (0.31-0.80) vs. 0.18 (0.01-0.60), p<0.05], and rectum [0.81 (0.37-0.91) vs. 0.57 (0.31-0.77), p<0.01]). The median contouring times were 2.6 (sMM) and 1.3min (sSM).

Multi-Atlas Image Soft Segmentation via Computation of the Expected Label Value.

The use of multiple atlases is common in medical image segmentation. This typically requires deformable registration of the atlases (or the average atlas) to the new image, which is computationally expensive and susceptible to entrapment in local optima. We propose to instead consider the probability of all possible atlas-to-image transformations and compute the expected label value (ELV), thereby not relying merely on the transformation deemed “optimal” by the registration method. Moreover, we do so without actually performing deformable registration, thus avoiding the associated computational costs. We evaluate our ELV computation approach by applying it to brain, liver, and pancreas segmentation on datasets of magnetic resonance and computed tomography images.

A stony coral cell atlas illuminates the molecular and cellular basis of coral symbiosis, calcification, and immunity

SummaryStony corals are colonial cnidarians that sustain the most biodiverse marine ecosystems on Earth: coral reefs. Despite their ecological importance, little is known about the cell types and molecular pathways that underpin the biology of reef-building corals. Using single-cell RNA sequencing, we define over 40 cell types across the life cycle of Stylophora pistillata. We discover specialized immune cells, and we uncover the developmental gene expression dynamics of calcium-carbonate skeleton formation. By simultaneously measuring the transcriptomes of coral cells and the algae within them, we characterize the metabolic programs involved in symbiosis in both partners. We also trace the evolution of these coral cell specializations by phylogenetic integration of multiple cnidarian cell type atlases. Overall, this study reveals the molecular and cellular basis of stony coral biology.

The Effects of Accelerated Brain Aging on the Antidepressant Response in Late-Life Depression

Introduction Late-Life Depression (LLD) is characterized by functional and structural brain alterations and accelerated brain aging. Our past work utilized a brain-based marker of accelerated aging to examine LLD and demonstrated a cross-sectional association between accelerated brain aging and impaired cognitive ability (poor episodic memory, slow processing speed, and poor executive function) as well as disability (measured by WHODAS). However, it is unclear whether or not accelerated brain aging influences the antidepressant response in depressed elders. Methods We examined depressed adults over 60 years-old with current Major Depressive Disorder (MDD) without psychotic features (N=95) randomized to receive either escitalopram up to 20mg daily or identical placebo (2:1 subject ratio) for 8 weeks. Subjects underwent baseline depression assessment (MADRS) and MRI prior to randomization. After using an automated deep learning tool to estimate brain age from cranial MRI, the brain age gap (BAG) was calculated by subtracting the participant's chronological age from his or her calculated brain age estimation. The brain MRI estimating tool uses ROI volumes derived from multiple atlases and the BrainAGE algorithm - the algorithm applies a deep convolutional neural network regression model trained on 5000 healthy controls. For statistical analysis, the dependent outcome variable was change in MADRS over 8 weeks. Initial models examined the effects of BAG with baseline MADRS, age, sex, and treatment assignment as covariates. Subsequent models tested for a statistical interaction between treatment assignment and BAG to observe whether or not the effects of brain aging on change in MADRS differed between arms. Results In the primary effects model examining change in MADRS, although there was a strong effect of treatment assignment (t = -3.75, p = 0.0003), there was no significant relationship between BAG and antidepressant response (t = -0.02, p = 0.9819). In the subsequent model, we observed a statistical interaction between BAG and treatment assignment on change in MADRS (t = -2.38, p = 0.0196). Analysis of subgroup data demonstrated that a larger BAG (indicating an older appearing brain) was associated with less change in MADRS for those who received placebo. In contrast, a larger BAG (indicating an older appearing brain) was associated with greater change in MADRS for those who received escitalopram. Conclusions The effect of accelerated brain aging, measured with the BAG, on change in MADRS have opposite patterns based on treatment assignment. Although a larger BAG is associated with poorer cognitive performance and greater disability, a larger BAG was associated with greater change in MADRS in the active treatment arm of our study. This is somewhat discrepant with prior work that associates poor cognitive performance with worse antidepressant response. As a global brain metric, BAG is not specific to discrete functional and/or structural brain impairments. Therefore, the elements that jointly contribute to the brain age metric may act in discrete, separate ways, with some influencing cognitive performance, and with others having less of a functional impact, serving as a marker of positive response to antidepressant medication. Further work must examine the complex relationships between accelerated brain aging, cognition, and response to antidepressants. Funding R01 MH102246, K24 MH110598

Multiatlas‐based segmentation of female pelvic organs: Application for computer‐aided diagnosis of cervical cancer

AbstractAtlas‐based segmentation is a high level segmentation technique which has become a standard paradigm for exploiting prior knowledge in image segmentation. Recent multiatlas‐based methods have provided greatly accurate segmentations of different parts of the human body by propagating manual delineations from multiple atlases in a data set to a query subject and fusing them. The female pelvic region is known to be of high variability which makes the segmentation task difficult. We propose, here, an approach for the segmentation of magnetic resonance imaging (MRI) called multiatlas‐based segmentation using online machine learning (OML). The proposed approach allows separating regions which may be affected by cervical cancer in a female pelvic MRI. The suggested approach is based on an online learning method for the construction of the dataset of atlases. The experiments demonstrate the higher accuracy of the suggested approach compared to a segmentation technique based on a fixed dataset of atlases and single‐atlas‐based segmentation technique.

A comparison of Freesurfer and multi-atlas MUSE for brain anatomy segmentation: Findings about size and age bias, and inter-scanner stability in multi-site aging studies

Automatic segmentation of brain anatomy has been a key processing step in quantitative neuroimaging analyses. An extensive body of literature has relied on Freesurfer segmentations. Yet, in recent years, the multi-atlas segmentation framework has consistently obtained results with superior accuracy in various evaluations. We compared brain anatomy segmentations from Freesurfer, which uses a single probabilistic atlas strategy, against segmentations from Multi-atlas region Segmentation utilizing Ensembles of registration algorithms and parameters and locally optimal atlas selection (MUSE), one of the leading ensemble-based methods that calculates a consensus segmentation through fusion of anatomical labels from multiple atlases and registrations. The focus of our evaluation was twofold. First, using manual ground-truth hippocampus segmentations, we found that Freesurfer segmentations showed a bias towards over-segmentation of larger hippocampi, and under-segmentation in older age. This bias was more pronounced in Freesurfer-v5.3, which has been used in multiple previous studies of aging, while the effect was mitigated in more recent Freesurfer-v6.0, albeit still present. Second, we evaluated inter-scanner segmentation stability using same day scan pairs from ADNI acquired on 1.5T and 3T scanners. We also found that MUSE obtains more consistent segmentations across scanners compared to Freesurfer, particularly in the deep structures.

Human Brain Atlases in Stroke Management

Stroke is a leading cause of death and a major cause of permanent disability. Its management is demanding because of variety of protocols, imaging modalities, pulse sequences, hemodynamic maps, criteria for treatment, and time constraints to promptly evaluate and treat. To cope with some of these issues, we propose novel, patented solutions in stroke management by employing multiple brain atlases for diagnosis, treatment, and prediction. Numerous and diverse CT and MRI scans are used: ARIC cohort, ischemic and hemorrhagic stroke CT cases, MRI cases with multiple pulse sequences, and 128 stroke CT patients, each with 170 variables and one year follow-up. The method employs brain atlases of anatomy, blood supply territories, and probabilistic stroke atlas. It rapidly maps an atlas to scan and provides atlas-assisted scan processing. Atlas-to-scan mapping is application-dependent and handles three types of regions of interest (ROIs): atlas-defined ROIs, atlas-quantified ROIs, and ROIs creating an atlas. An ROI is defined by atlas-guided anatomy or scan-derived pathology. The atlas defines ROI or quantifies it. A brain atlas potential has been illustrated in four atlas-assisted applications for stroke occurrence prediction and screening, rapid and automatic stroke diagnosis in emergency room, quantitative decision support in thrombolysis in ischemic stroke, and stroke outcome prediction and treatment assessment. The use of brain atlases in stroke has many potential advantages, including rapid processing, automated and robust handling, wide range of applications, and quantitative assessment. Further work is needed to enhance the developed prototypes, clinically validate proposed solutions, and introduce them to clinical practice.

Resolving single-cell heterogeneity from hundreds of thousands of cells through sequential hybrid clustering and NMF.

MotivationThe rapid proliferation of single-cell RNA-sequencing (scRNA-Seq) technologies has spurred the development of diverse computational approaches to detect transcriptionally coherent populations. While the complexity of the algorithms for detecting heterogeneity has increased, most require significant user-tuning, are heavily reliant on dimension reduction techniques and are not scalable to ultra-large datasets. We previously described a multi-step algorithm, Iterative Clustering and Guide-gene Selection (ICGS), which applies intra-gene correlation and hybrid clustering to uniquely resolve novel transcriptionally coherent cell populations from an intuitive graphical user interface.ResultsWe describe a new iteration of ICGS that outperforms state-of-the-art scRNA-Seq detection workflows when applied to well-established benchmarks. This approach combines multiple complementary subtype detection methods (HOPACH, sparse non-negative matrix factorization, cluster ‘fitness’, support vector machine) to resolve rare and common cell-states, while minimizing differences due to donor or batch effects. Using data from multiple cell atlases, we show that the PageRank algorithm effectively downsamples ultra-large scRNA-Seq datasets, without losing extremely rare or transcriptionally similar yet distinct cell types and while recovering novel transcriptionally distinct cell populations. We believe this new approach holds tremendous promise in reproducibly resolving hidden cell populations in complex datasets.Availability and implementationICGS2 is implemented in Python. The source code and documentation are available at http://altanalyze.org.Supplementary information Supplementary data are available at Bioinformatics online.

Diffeomorphic Registration With Intensity Transformation and Missing Data: Application to 3D Digital Pathology of Alzheimer's Disease.

This paper examines the problem of diffeomorphic image registration in the presence of differing image intensity profiles and sparsely sampled, missing, or damaged tissue. Our motivation comes from the problem of aligning 3D brain MRI with 100-micron isotropic resolution to histology sections at 1 × 1 × 1,000-micron resolution with multiple varying stains. We pose registration as a penalized Bayesian estimation, exploiting statistical models of image formation where the target images are modeled as sparse and noisy observations of the atlas. In this injective setting, there is no assumption of symmetry between atlas and target. Cross-modality image matching is achieved by jointly estimating polynomial transformations of the atlas intensity. Missing data is accommodated via a multiple atlas selection procedure where several atlas images may be of homogeneous intensity and correspond to “background” or “artifact.” The two concepts are combined within an Expectation-Maximization algorithm, where atlas selection posteriors and deformation parameters are updated iteratively and polynomial coefficients are computed in closed form. We validate our method with simulated images, examples from neuropathology, and a standard benchmarking dataset. Finally, we apply it to reconstructing digital pathology and MRI in standard atlas coordinates. By using a standard convolutional neural network to detect tau tangles in histology slices, this registration method enabled us to quantify the 3D density distribution of tauopathy throughout the medial temporal lobe of an Alzheimer's disease postmortem specimen.

Adaptive sparse learning using multi-template for neurodegenerative disease diagnosis

Neurodegenerative diseases are excessively affecting millions of patients, especially elderly people. Early detection and management of these diseases are crucial as the clinical symptoms take years to appear after the onset of neuro-degeneration. This paper proposes an adaptive feature learning framework using multiple templates for early diagnosis. A multi-classification scheme is developed based on multiple brain parcellation atlases with various regions of interest. Different sets of features are extracted and then fused, and a feature selection is applied with an adaptively chosen sparse degree. In addition, both linear discriminative analysis and locally preserving projections are integrated to construct a least square regression model. Finally, we propose a feature space to predict the severity of the disease by the guidance of clinical scores. Our proposed method is validated on both Alzheimer's disease neuroimaging initiative and Parkinson's progression markers initiative databases. Extensive experimental results suggest that the proposed method outperforms the state-of-the-art methods, such as the multi-modal multi-task learning or joint sparse learning. Our method demonstrates that accurate feature learning facilitates the identification of the highly relevant brain regions with significant contribution in the prediction of disease progression. This may pave the way for further medical analysis and diagnosis in practical applications.

FCN Based Label Correction for Multi-Atlas Guided Organ Segmentation.

Segmentation of medical images using multiple atlases has recently gained immense attention due to their augmented robustness against variabilities across different subjects. These atlas-based methods typically comprise of three steps: atlas selection, image registration, and finally label fusion. Image registration is one of the core steps in this process, accuracy of which directly affects the final labeling performance. However, due to inter-subject anatomical variations, registration errors are inevitable. The aim of this paper is to develop a deep learning-based confidence estimation method to alleviate the potential effects of registration errors. We first propose a fully convolutional network (FCN) with residual connections to learn the relationship between the image patch pair (i.e., patches from the target subject and the atlas) and the related label confidence patch. With the obtained label confidence patch, we can identify the potential errors in the warped atlas labels and correct them. Then, we use two label fusion methods to fuse the corrected atlas labels. The proposed methods are validated on a publicly available dataset for hippocampus segmentation. Experimental results demonstrate that our proposed methods outperform the state-of-the-art segmentation methods.

Multiatlas Calibration of Biophysical Brain Tumor Growth Models with Mass Effect.

We present a 3D fully-automatic method for the calibration of partial differential equation (PDE) models of glioblastoma (GBM) growth with "mass effect", the deformation of brain tissue due to the tumor. We quantify the mass effect, tumor proliferation, tumor migration, and the localized tumor initial condition from a single multiparameteric Magnetic Resonance Imaging (mpMRI) patient scan. The PDE is a reaction-advection-diffusion partial differential equation coupled with linear elasticity equations to capture mass effect. The single-scan calibration model is notoriously difficult because the precancerous (healthy) brain anatomy is unknown. To solve this inherently ill-posed and illconditioned optimization problem, we introduce a novel inversion scheme that uses multiple brain atlases as proxies for the healthy precancer patient brain resulting in robust and reliable parameter estimation. We apply our method on both synthetic and clinical datasets representative of the heterogeneous spatial landscape typically observed in glioblastomas to demonstrate the validity and performance of our methods. In the synthetic data, we report calibration errors (due to the ill-posedness and our solution scheme) in the 10%-20% range. In the clinical data, we report good quantitative agreement with the observed tumor and qualitative agreement with the mass effect (for which we do not have a ground truth). Our method uses a minimal set of parameters and provides both global and local quantitative measures of tumor infiltration and mass effect.

Optimal Target Delineation and Treatment Techniques in the Era of Conformal Photon and Proton Breast and Regional Nodal Irradiation

PurposeRegional nodal irradiation improves disease-free and distant disease–free survival in patients with high-risk breast cancer (BC). Trials demonstrating this used 2- or 3-dimensional conformal radiation therapy (2-dimensional or 3-dimensional [3D] conformal radiotherapy [CRT]) fields based on bony anatomy. Modern volumetric-modulated arc therapy (VMAT) and pencil beam scanning proton therapy (PBSPT) may underdose regional nodes (RNs) not contoured but covered by 3D CRT. Multiple atlases guide modern treatment planning. This study addresses the risk of underdosing when relying on published atlases and treating with 3D CRT, VMAT, and PBSPT. Methods and MaterialsTargets per the Radiation Therapy Oncology Group (RTOG), European Society for Radiotherapy and Oncology (ESTRO), and Radiotherapy Comparative Effectiveness Consortium (RADCOMP) atlases were contoured on a representative patient CT scan. 3D CRT plans based on anatomic borders and VMAT and PBSPT plans for each set of target volumes were generated. Positron emission tomography/computed tomography (PET/CT) scans were reviewed. CT-positive and 18F-fluorodeoxyglucose (18F-FDG)–avid RNs (n = 389) were mapped from 102 patients with locally advanced (n = 51; median 2; range, 1-8 nodes) and metastatic (n = 51; median 4; range, 1-19 nodes) BC: axillary (AX; n = 284), supraclavicular (SCV; n = 60), and internal mammary nodal (IMN; n = 45). 18F-FDG-avid RNs falling within the 95% isodose line were considered adequately covered. Results3D CRT plans provided excellent RN coverage. Low AX nodes were covered (≥99%) in all plans. Underdosing of 18F-FDG–avid RNs falling in the high AX (78%-92%), SCV (52%-75%), and IMN (84%-89%) volumes was observed following the RTOG and ESTRO atlases for VMAT and PBSPT plans. Use of the RADCOMP atlas provided coverage of these areas (89%-100%) with slightly increased heart and lung doses. Atlas guided VMAT/PBSPT plans provided cumulative nodal coverage as follows: ESTRO (89%/88%), RTOG (93%/91%), and RADCOMP (98%/96%). ConclusionsVMAT and PBSPT for regional nodal irradiation in patients with high-risk BC risks underdosage in the high AX, SCV, and IMN nodal regions unless comprehensive target delineation is performed.

High-order Feature Learning for Multi-atlas based Label Fusion: Application to Brain Segmentation with MRI.

Multi-atlas based segmentation methods have shown their effectiveness in brain regions-of-interesting (ROIs) segmentation, by propagating labels from multiple atlases to a target image based on the similarity between patches in the target image and multiple atlas images. Most of the existing multiatlas based methods use image intensity features to calculate the similarity between a pair of image patches for label fusion. In particular, using only low-level image intensity features cannot adequately characterize the complex appearance patterns (e.g., the high-order relationship between voxels within a patch) of brain magnetic resonance (MR) images. To address this issue, this paper develops a high-order feature learning framework for multi-atlas based label fusion, where high-order features of image patches are extracted and fused for segmenting ROIs of structural brain MR images. Specifically, an unsupervised feature learning method (i.e., means-covariances restricted Boltzmann machine, mcRBM) is employed to learn high-order features (i.e., mean and covariance features) of patches in brain MR images. Then, a group-fused sparsity dictionary learning method is proposed to jointly calculate the voting weights for label fusion, based on the learned high-order and the original image intensity features. The proposed method is compared with several state-of-the-art label fusion methods on ADNI, NIREP and LONI-LPBA40 datasets. The Dice ratio achieved by our method is 88:30%, 88:83%, 79:54% and 81:02% on left and right hippocampus on the ADNI, NIREP and LONI-LPBA40 datasets, respectively, while the best Dice ratio yielded by the other methods are 86:51%, 87:39%, 78:48% and 79:65% on three datasets, respectively.

Diffusion Tensor Image segmentation based on multi-atlas Active Shape Model

Active Shape Model (ASM) has been successfully applied in the segmentation of Diffusion Tensor Magnetic Resonance Image (DT-MRI, referred to as DTI) of brain. However, due to multiple anatomical structure types, irregular shapes, small gray-scale and large amount of these images, perfect segmentation performance could not be achieved. Especially, it is sensitive to initial values with high computational complexity. In this paper, we introduce the gray information of multiple atlases and the prior information of target shapes into the ASM and propose the Multi-Atlas Active Shape Model (referred to as MA-ASM) approach for DTI segmentation. It was evaluated in a manually labeled database with 7 Region of Interest (ROI)s for each of 20 subjects. In comparison with the state of art method of STAPLE (Simultaneous Truth Performance Level Estimation), the proposed algorithm was closer to the manual segmentation shape by subjective visual effects, and had higher overlap rates and lower error detection rates on quantitative analysis than STAPLE.

Simultaneous Tissue Classification and Lateral Ventricle Segmentation via a 2D U-net Driven by a 3D Fully Convolutional Neural Network.

In this paper, we proposed and validated a novel and fully automatic pipeline for simultaneous tissue classification and lateral ventricle segmentation via a 2D U-net. The 2D U-net was driven by a 3D fully convolutional neural network (FCN). Multiple T1-weighted atlases which had been pre-segmented into six whole-brain regions including the gray matter (GM), white matter (WM), cerebrospinal fluid (CSF), lateral ventricles (LVs), skull, and the background of the entire image were used. In the proposed pipeline, probability maps of the six whole-brain regions of interest (ROIs) were obtained after a pre-segmentation through a trained 3D patch-based FCN. To further capture the global context of the entire image, the to-be-segmented image and the corresponding six probability maps were input to a trained 2D U-net in a 2D slice fashion to obtain the final segmentation map. Experiments were performed on a dataset consisting of 18 T1-weighted images. Compared to the 3D patch-based FCN on segmenting five ROIs (GM, WM, CSF, LVs, skull) and another two classical methods (SPM and FSL) on segmenting GM and WM, the proposed pipeline showed a superior segmentation performance. The proposed segmentation architecture can also be extended to other medical image segmentation tasks.

Expected Label Value Computation for Atlas-Based Image Segmentation.

The use of multiple atlases is common in medical image segmentation. This typically requires deformable registration of the atlases (or the average atlas) to the new image, which is computationally expensive and susceptible to entrapment in local optima. We propose to instead consider the probability of all possible transformations and compute the expected label value (ELV), thereby not relying merely on the transformation resulting from the registration. Moreover, we do so without actually performing deformable registration, thus avoiding the associated computational costs. We evaluate our ELV computation approach by applying it to liver segmentation on a dataset of computed tomography (CT) images.

Multiseg pipeline: automatic tissue segmentation of brain MR images with subject-specific atlases.

Automated segmentation and labeling of individual brain anatomical regions is challenging due to individual structural variability. Although, atlas-based segmentation has shown its potential for both tissue and structure segmentation, the inherent natural variability as well as disease-related changes in MR appearance is often inappropriately represented by a single atlas image. In order to have a more accurate representation, several atlases may be used for the segmentation task in a given neuroimaging study. In this paper, we present the MultisegPipeline, it uses multiple atlases that have been visually inspected and capture the expected variability in a neonatal population. The MultisegPipeline transfers the labeled regions from each atlas to the target image using deformable registration (ANTs1 or QuickSilver2 is available for this task). Additionally, the set of labels are merged using a label fusion technique that reduces the errors produced by the registration. The final output is a single label map that combines the results produced by all atlases into a consensus solution. In our study, the MultisegPipeline is used to segment brain MR images from 31 infants, a leave-one-out strategy was used to test our framework. The average dice score coefficient was 0.89.