Multi-atlas segmentation with robust label transfer and label fusion

Characterization of Atypical Corticospinal Tract Microstructure and Hand Impairments in Early-Onset Hemiplegic Cerebral Palsy: Preliminary Findings

Unilateral brain injuries occurring before at or shortly after full-term can result in hemiplegic cerebral palsy (HCP). HCP affects one side of the body and can be characterized in the hand with measures of weakness and a loss of independent hand control resulting in mirror movements. Hand impairment severity is extremely heterogeneous across individuals with HCP and the neural basis for this variability is unclear. We used diffusion MRI and tractography to investigate the relationship between structural morphology of the supraspinal corticospinal tract (CST) and the severity of two typical hand impairments experienced by individuals with HCP, grasp weakness and mirror movements. Results from nine children with HCP and eight children with typical development show that there is a significant hemispheric association between CST microstructure and hand impairment severity that may be explained by atypical development and fiber distribution of motor pathways. Further analysis in the non-lesioned (dominant) hemisphere shows significant differences for CST termination in the cortex between participants with HCP and those with typical development. These findings suggest that structural disparities at the cellular level in the seemingly unaffected hemisphere after early unilateral brain injury may be the cause of heterogeneous hand impairments seen in this population.Clinical Relevance- Quantitative measurement of the variability in hand function in individuals with HCP is necessary to represent the distinct impairments experienced by each person. Further understanding of the structural neural morphology underlying distal upper extremity motor deficits after early unilateral brain injury will help lead to the development of more specific targeted interventions that increase functional outcomes.

Functional and Structural Brain Connectivity in Children With Bilateral Cerebral Palsy Compared to Age-Related Controls and in Response to Intensive Rapid-Reciprocal Leg Training

Background

Compared to unilateral cerebral palsy (CP), less is known about brain reorganization and plasticity in bilateral CP especially in relation or response to motor training. The few trials that reported brain imaging results alongside functional outcomes include a handful of studies in unilateral CP, and one pilot trial of three children with bilateral CP. This study is the first locomotor training randomized controlled trial (RCT) in bilateral CP to our knowledge reporting brain imaging outcomes.

Methods

Objective was to compare MRI brain volumes, resting state connectivity and white matter integrity using DTI in children with bilateral CP with PVL and preterm birth history (<34 weeks), to age-related controls, and from an RCT of intensive 12 week rapid-reciprocal locomotor training using an elliptical or motor-assisted cycle. We hypothesized that connectivity in CP compared to controls would be greater across sensorimotor-related brain regions and that functional (resting state) and structural (fractional anisotropy) connectivity would improve post intervention. We further anticipated that baseline and post-intervention imaging and functional measures would correlate.

Results

Images were acquired with a 3T MRI scanner for 16/27 children with CP in the trial, and 18 controls. No conclusive evidence of training-induced neuroplastic effects were seen. However, analysis of shared variance revealed that greater increases in precentral gyrus connectivity with the thalamus and pons may be associated with larger improvements in the trained device speed. Exploratory analyses also revealed interesting potential relationships between brain integrity and multiple functional outcomes in CP, with functional connectivity between the motor cortex and midbrain showing the strongest potential relationship with mobility. Decreased posterior white matter, corpus callosum and thalamic volumes, and FA in the posterior thalamic radiation were the most prominent group differences with corticospinal tract differences notably not found.

Conclusions

Results reinforce the involvement of sensory-related brain areas in bilateral CP. Given the wide individual variability in imaging results and clinical responses to training, a greater focus on neural and other mechanisms related to better or worse outcomes is recommended to enhance rehabilitation results on a patient vs. group level.

Clinical and Molecular Correlates of Abnormal Changes in the Cerebellum and Globus Pallidus in Fragile X Premutation

BACKGROUND

Fragile X premutation carriers (55-200 CGG triplets) may develop a progressive neurodegenerative disorder, fragile X-associated tremor/ataxia syndrome (FXTAS), after the age of 50. The neuroradiologic markers of FXTAS are hyperintense T2-signals in the middle cerebellar peduncle-the MCP sign. We recently noticed abnormal T2-signals in the globus pallidus in male premutation carriers and controls but the prevalence and clinical significance were unknown.

METHODS

We estimated the prevalence of the MCP sign and pallidal T2-abnormalities in 230 male premutation carriers and 144 controls (aged 8-86), and examined the associations with FXTAS symptoms, CGG repeat length, and iron content in the cerebellar dentate nucleus and globus pallidus.

RESULTS

Among participants aged ≥45 years (175 premutation carriers and 82 controls), MCP sign was observed only in premutation carriers (52 vs. 0%) whereas the prevalence of pallidal T2-abnormalities approached significance in premutation carriers compared with controls after age-adjustment (25.1 vs. 13.4%, p = 0.069). MCP sign was associated with impaired motor and executive functioning, and the additional presence of pallidal T2-abnormalities was associated with greater impaired executive functioning. Among premutation carriers, significant iron accumulation was observed in the dentate nucleus, and neither pallidal or MCP T2-abnormalities affected measures of the dentate nucleus. While the MCP sign was associated with CGG repeat length >75 and dentate nucleus volume correlated negatively with CGG repeat length, pallidal T2-abnormalities did not correlate with CGG repeat length. However, pallidal signal changes were associated with age-related accelerated iron depletion and variability and having both MCP and pallidal signs further increased iron variability in the globus pallidus.

CONCLUSIONS

Only the MCP sign, not pallidal abnormalities, revealed independent associations with motor and cognitive impairment; however, the occurrence of combined MCP and pallidal T2-abnormalities may present a risk for greater cognitive impairment and increased iron variability in the globus pallidus.

A 7 Tesla Amygdalar-Hippocampal Shape Analysis of Lithium Response in Bipolar Disorder

Research to discover clinically useful predictors of lithium response in patients with bipolar disorder has largely found them to be elusive. We demonstrate here that detailed neuroimaging may have the potential to fill this important gap in mood disorder therapeutics. Lithium treatment and bipolar disorder have both been shown to affect anatomy of the hippocampi and amygdalae but there is no consensus on the nature of their effects. We aimed to investigate structural surface anatomy changes in amygdala and hippocampus correlated with treatment response in bipolar disorder. Patients with bipolar disorder (N = 14) underwent lithium treatment, were classified by response status at acute and long-term time points, and scanned with 7 Tesla structural MRI. Large Deformation Diffeomorphic Metric Mapping was applied to detect local differences in hippocampal and amygdalar anatomy between lithium responders and non-responders. Anatomy was also compared to 21 healthy comparison participants. A patch of the ventral surface of the left hippocampus was found to be significantly atrophied in non-responders as compared to responders at the acute time point and was associated at a trend-level with long-term response status. We did not detect an association between response status and surface anatomy of the right hippocampus or amygdala. To the best of our knowledge, this is the first shape analysis of hippocampus and amygdala in bipolar disorder using 7 Tesla MRI. These results can inform future work investigating possible neuroimaging predictors of lithium response in bipolar disorder.

Whole brain volume and cortical thickness abnormalities in Wilson's disease: a clinical correlation study

Wilson's disease (WD) is an inherited autosomal recessive disorder of copper metabolism, and its neurological and neuropsychiatric manifestations are associated with copper accumulation in brain. A few neuroimaging studies have shown that gray matter atrophy in WD affects both subcortical structures and cortex. This study aims to quantitatively evaluate the morphometric brain abnormalities in patients with WD in terms of whole brain volume and cortical thickness and their associations with clinical severity of WD. Thirty patients clinically diagnosed as WD with neurological manifestations and 25 healthy controls (HC) were recruited. 3D T1-weighted images were segmented into 276 whole-brain regions of interest (ROIs) and 68 cortical ROIs. WD-vs-HC group comparisons were then conducted for each ROI. The associations between those morphometric measurements and the Global Assessment Scale (GAS) score for WD were analyzed. Compared with HC, significant WD-related volumetric decreases were found in the bilateral subcortical nuclei (putamen, globus pallidus, caudate nucleus, substantia nigra, red nucleus and thalamus), diffuse white matter and several gray matter regions. WD patients showed reduced cortical thickness in the left precentral gyrus and the left insula. Further, the volumes of the right globus pallidus, bilateral putamen, right external capsule and left superior longitudinal fasciculus were negatively correlated with GAS. Our results indicated that significant WD-related morphometric abnormalities were quantified in terms of whole-brain volumes and cortical thicknesses, some of which correlated significantly to the clinical severity of WD. Those morphometrics may provide a potentially effective biomarker of WD.

Interaction between ventricular expansion and structural changes in the corpus callosum and putamen in males with FMR1 normal and premutation alleles

Ventricular enlargement (VE) is commonly observed in aging and fragile X-associated tremor/ataxia syndrome (FXTAS), a late-onset neurodegenerative disorder. VE may generate a mechanical force causing structural deformation. In this longitudinal study, we examined the relationships between VE and structural changes in the corpus callosum (CC) and putamen. MRI scans (2-7/person over 0.2-7.5 years) were acquired from 22 healthy controls, 26 unaffected premutation carriers (PFX-), and 39 carriers affected with FXTAS (PFX+). Compared with controls, PFX- demonstrated enlarged fourth ventricles, whereas PFX+ displayed enlargement in both third and fourth ventricles, CC thinning, putamen atrophy/deformation (thinning and increased distance), and accelerated expansions in lateral ventricles. Common for all groups, baseline VE predicted accelerated CC thinning and putamen atrophy/deformation and conversely, baseline CC and putamen atrophy/deformation and enlarged third and fourth ventricles predicted accelerated lateral ventricular expansion. The results suggest a progressive VE within the 4 ventricles as FXTAS develops and a deleterious cycle between VE and brain deformation that may commonly occur during aging and FXTAS progression but become accelerated in FXTAS.

Developmental aspects of FXAND in a man with the FMR1 premutation

BACKGROUND

Fragile X mental retardation 1 (FMR1) premutation can cause developmental problems including autism spectrum disorder (ASD), social anxiety, depression, and attention deficit hyperactivity disorder (ADHD). These problems fall under an umbrella term of Fragile X-associated Neuropsychiatric Disorders (FXAND) and is separate from Fragile X-associated Tremor/Ataxia syndrome (FXTAS), a neurodegenerative disorder.

METHODS/CLINICAL CASE

A 26-year-old Caucasian male with the Fragile X premutation who presented with multiple behavior and emotional problems including depression and anxiety at 10 years of age. He was evaluated at 13, 18, and 26 years old with age-appropriate cognitive assessments, psychiatric evaluations, and an MRI of the brain.

RESULTS

The Autism Diagnostic Observation Scale (ADOS) was done at 13 years old and showed the patient has autism spectrum disorder (ASD). An evaluation at 18 years old showed a full-scale IQ of 64. A Kiddie Schedule for Affective Disorders and Schizophrenia (K-SADS) performed at 26 years old confirmed the previous impression of social anxiety disorder, agoraphobia disorder, and selective mutism. His MRI acquired at 26 years old showed enlarged ventricles, increased frontal subarachnoid spaces, and hypergyrification.

CONCLUSION

This is an exemplary case of an FMR1 premutation carrier with significant psychiatric and cognitive issues that demonstrates Fragile X-associated Neuropsychiatric Disorders (FXAND) as separate from the other well-known premutation disorders.

Linear Registration of Brain MRI Using Knowledge-Based Multiple Intermediator Libraries

Linear registration is often the crucial first step for various types of image analysis. Although this is mathematically simple, failure is not uncommon. When investigating the brain by magnetic resonance imaging (MRI), the brain is the target organ for registration but the existence of other tissues, in addition to a variety of fields of view, different brain locations, orientations and anatomical features, poses some serious fundamental challenges. Consequently, a number of different algorithms have been put forward to minimize potential errors. In the present study, we tested a knowledge-based approach that can be combined with any form of registration algorithm. This approach consisted of a library of intermediate images (mediators) with known transformation to the target image. Test images were first registered to all mediators and the best mediator was selected to ensure optimum registration to the target. In order to select the best mediator, we evaluated two similarity criteria: the sum of squared differences and mutual information. This approach was applied to 48 mediators and 96 test images. In order to reduce one of the main drawbacks of the approach, increased computation time, we reduced the size of the library by clustering. Our results indicated clear improvement in registration accuracy.

Automated segmentation of medial temporal lobe subregions on in vivo T1-weighted MRI in early stages of Alzheimer's disease

Medial temporal lobe (MTL) substructures are the earliest regions affected by neurofibrillary tangle pathology-and thus are promising biomarkers for Alzheimer's disease (AD). However, automatic segmentation of the MTL using only T1-weighted (T1w) magnetic resonance imaging (MRI) is challenging due to the large anatomical variability of the MTL cortex and the confound of the dura mater, which is commonly segmented as gray matter by state-of-the-art algorithms because they have similar intensity in T1w MRI. To address these challenges, we developed a novel atlas set, consisting of 15 cognitively normal older adults and 14 patients with mild cognitive impairment with a label explicitly assigned to the dura, that can be used by the multiatlas automated pipeline (Automatic Segmentation of Hippocampal Subfields [ASHS-T1]) for the segmentation of MTL subregions, including anterior/posterior hippocampus, entorhinal cortex (ERC), Brodmann areas (BA) 35 and 36, and parahippocampal cortex on T1w MRI. Cross-validation experiments indicated good segmentation accuracy of ASHS-T1 and that the dura can be reliably separated from the cortex (6.5% mislabeled as gray matter). Conversely, FreeSurfer segmented majority of the dura mater (62.4%) as gray matter and the degree of dura mislabeling decreased with increasing disease severity. To evaluate its clinical utility, we applied the pipeline to T1w images of 663 ADNI subjects and significant volume/thickness loss is observed in BA35, ERC, and posterior hippocampus in early prodromal AD and all subregions at later stages. As such, the publicly available new atlas and ASHS-T1 could have important utility in the early diagnosis and monitoring of AD and enhancing brain-behavior studies of these regions.

A Multi-Atlas Label Fusion Tool for Neonatal Brain MRI Parcellation and Quantification

Structure-by-structure analysis, in which the brain magnetic resonance imaging (MRI) is parcellated based on its anatomical units, is widely used to investigate chronological changes in morphology or signal intensity during normal development, as well as to identify the alterations seen in various diseases or conditions. The multi-atlas label fusion (MALF) method is considered a highly accurate parcellation approach, and anticipated for clinical application to quantitatively evaluate early developmental processes. However, the current MALF methods, which are designed for neonatal brain segmentations, are not widely available. In this study, we developed a T1-weighted, neonatal, multi-atlas repository and integrated it into the MALF-based brain segmentation tools in the cloud-based platform, MRICloud. The cloud platform ensures users instant access to the advanced MALF tool for neonatal brains, with no software or installation requirements for the client. The Web platform by braingps.mricloud.org will eliminate the dependence on a particular operating system (eg, Windows, Macintosh, or Linux) and the requirement for high computational performance of the user's computers. The MALF-based, fully automated, image parcellation could achieve excellent agreement with manual parcellation, and the whole and regional brain volumes quantified through this method demonstrated developmental trajectories comparable to those from a previous publication. This solution will make the latest MALF tools readily available to users, with minimum barriers, and will expedite and accelerate advancements in developmental neuroscience research, neonatology, and pediatric neuroradiology.

Automated Generation of Radiologic Descriptions on Brain Volume Changes From T1-Weighted MR Images: Initial Assessment of Feasibility

Purpose: To examine the feasibility and potential difficulties of automatically generating radiologic reports (RRs) to articulate the clinically important features of brain magnetic resonance (MR) images.

Materials and Methods: We focused on examining brain atrophy by using magnetization-prepared rapid gradient-echo (MPRAGE) images. The technology was based on multi-atlas whole-brain segmentation that identified 283 structures, from which larger superstructures were created to represent the anatomic units most frequently used in RRs. Through two layers of data-reduction filters, based on anatomic and clinical knowledge, raw images (~10 MB) were converted to a few kilobytes of human-readable sentences. The tool was applied to images from 92 patients with memory problems, and the results were compared to RRs independently produced by three experienced radiologists. The mechanisms of disagreement were investigated to understand where machine–human interface succeeded or failed.

Results: The automatically generated sentences had low sensitivity (mean: 24.5%) and precision (mean: 24.9%) values; these were significantly lower than the inter-rater sensitivity (mean: 32.7%) and precision (mean: 32.2%) of the radiologists. The causes of disagreement were divided into six error categories: mismatch of anatomic definitions (7.2 ± 9.3%), data-reduction errors (11.4 ± 3.9%), translator errors (3.1 ± 3.1%), difference in the spatial extent of used anatomic terms (8.3 ± 6.7%), segmentation quality (9.8 ± 2.0%), and threshold for sentence-triggering (60.2 ± 16.3%).

Conclusion: These error mechanisms raise interesting questions about the potential of automated report generation and the quality of image reading by humans. The most significant discrepancy between the human and automatically generated RRs was caused by the sentence-triggering threshold (the degree of abnormality), which was fixed to z-score >2.0 for the automated generation, while the thresholds by radiologists varied among different anatomical structures.

Brain MRI Pattern Recognition Translated to Clinical Scenarios

We explored the performance of structure-based computational analysis in four neurodegenerative conditions [Ataxia (AT, n = 16), Huntington's Disease (HD, n = 52), Alzheimer's Disease (AD, n = 66), and Primary Progressive Aphasia (PPA, n = 50)], all characterized by brain atrophy. The independent variables were the volumes of 283 anatomical areas, derived from automated segmentation of T1-high resolution brain MRIs. The segmentation based volumetric quantification reduces image dimensionality from the voxel level [on the order of O(10⁶)] to anatomical structures [O(10²)] for subsequent statistical analysis. We evaluated the effectiveness of this approach on extracting anatomical features, already described by human experience and a priori biological knowledge, in specific scenarios: (1) when pathologies were relatively homogeneous, with evident image alterations (e.g., AT); (2) when the time course was highly correlated with the anatomical changes (e.g., HD), an analogy for prediction; (3) when the pathology embraced heterogeneous phenotypes (e.g., AD) so the classification was less efficient but, in compensation, anatomical and clinical information were less redundant; and (4) when the entity was composed of multiple subgroups that had some degree of anatomical representation (e.g., PPA), showing the potential of this method for the clustering of more homogeneous phenotypes that can be of clinical importance. Using the structure-based quantification and simple linear classifiers (partial least square), we achieve 87.5 and 73% of accuracy on differentiating AT and pre-symptomatic HD patents from controls, respectively. More importantly, the anatomical features automatically revealed by the classifiers agreed with the patterns previously described on these pathologies. The accuracy was lower (68%) on differentiating AD from controls, as AD does not display a clear anatomical phenotype. On the other hand, the method identified PPA clinical phenotypes and their respective anatomical signatures. Although most of the data are presented here as proof of concept in simulated clinical scenarios, structure-based analysis was potentially effective in characterizing phenotypes, retrieving relevant anatomical features, predicting prognosis, and aiding diagnosis, with the advantage of being easily translatable to clinics and understandable biologically.

Open-Label Allopregnanolone Treatment of Men with Fragile X-Associated Tremor/Ataxia Syndrome

Fragile X-associated tremor/ataxia syndrome (FXTAS) is a late-onset neurodegenerative disorder affecting approximately 45% of male and 16% of female carriers of the FMR1 premutation over the age of 50 years. Currently, no effective treatment is available. We performed an open-label intervention study to assess whether allopregnanolone, a neurosteroid promoting regeneration and repair, can improve clinical symptoms, brain activity, and magnetic resonance imaging (MRI) measurements in patients with FXTAS. Six patients underwent weekly intravenous infusions of allopregnanolone (2-6 mg over 30 min) for 12 weeks. All patients completed baseline and follow-up studies, though MRI scans were not collected from 1 patient because of MRI contraindications. The MRI scans from previous visits, along with scans from 8 age-matched male controls, were also included to establish patients' baseline condition as a reference. Functional outcomes included quantitative measurements of tremor and ataxia and neuropsychological evaluations. Brain activity consisted of event-related potential N400 word repetition effect during a semantic memory processing task. Structural MRI outcomes comprised volumes of the hippocampus, amygdala, and fluid-attenuated inversion recovery hyperintensities, and microstructural integrity of the corpus callosum. The results of the study showed that allopregnanolone infusions were well tolerated in all subjects. Before treatment, the patients disclosed impairment in executive function, verbal fluency and learning, and progressive deterioration of all MRI measurements. After treatment, the patients demonstrated improvement in executive functioning, episodic memory and learning, and increased N400 repetition effect amplitude. Although MRI changes were not significant as a group, both improved and deteriorated MRI measurements occurred in individual patients in contrast to uniform deterioration before the treatment. Significant correlations between baseline MRI measurements and changes in neuropsychological test scores indicated the effects of allopregnanolone on improving executive function, learning, and memory for patients with relatively preserved hippocampus and corpus callosum, while reducing psychological symptoms for patients with small hippocampi and amygdalae. The findings show the promise of allopregnanolone in improving cognitive functioning in patients with FXTAS and in partially alleviating some aspects of neurodegeneration. Further studies are needed to verify the efficacy of allopregnanolone for treating FXTAS.

A comparison of accurate automatic hippocampal segmentation methods

The hippocampus is one of the first brain structures affected by Alzheimer's disease (AD). While many automatic methods for hippocampal segmentation exist, few studies have compared them on the same data. In this study, we compare four fully automated hippocampal segmentation methods in terms of their conformity with manual segmentation and their ability to be used as an AD biomarker in clinical settings. We also apply error correction to the four automatic segmentation methods, and complete a comprehensive validation to investigate differences between the methods. The effect size and classification performance is measured for AD versus normal control (NC) groups and for stable mild cognitive impairment (sMCI) versus progressive mild cognitive impairment (pMCI) groups. Our study shows that the nonlinear patch-based segmentation method with error correction is the most accurate automatic segmentation method and yields the most conformity with manual segmentation (κ=0.894). The largest effect size between AD versus NC and sMCI versus pMCI is produced by FreeSurfer with error correction. We further show that, using only hippocampal volume, age, and sex as features, the area under the receiver operating characteristic curve reaches up to 0.8813 for AD versus NC and 0.6451 for sMCI versus pMCI. However, the automatic segmentation methods are not significantly different in their performance.

Manual-Protocol Inspired Technique for Improving Automated MR Image Segmentation during Label Fusion

Recent advances in multi-atlas based algorithms address many of the previous limitations in model-based and probabilistic segmentation methods. However, at the label fusion stage, a majority of algorithms focus primarily on optimizing weight-maps associated with the atlas library based on a theoretical objective function that approximates the segmentation error. In contrast, we propose a novel method—Autocorrecting Walks over Localized Markov Random Fields (AWoL-MRF)—that aims at mimicking the sequential process of manual segmentation, which is the gold-standard for virtually all the segmentation methods. AWoL-MRF begins with a set of candidate labels generated by a multi-atlas segmentation pipeline as an initial label distribution and refines low confidence regions based on a localized Markov random field (L-MRF) model using a novel sequential inference process (walks). We show that AWoL-MRF produces state-of-the-art results with superior accuracy and robustness with a small atlas library compared to existing methods. We validate the proposed approach by performing hippocampal segmentations on three independent datasets: (1) Alzheimer's Disease Neuroimaging Database (ADNI); (2) First Episode Psychosis patient cohort; and (3) A cohort of preterm neonates scanned early in life and at term-equivalent age. We assess the improvement in the performance qualitatively as well as quantitatively by comparing AWoL-MRF with majority vote, STAPLE, and Joint Label Fusion methods. AWoL-MRF reaches a maximum accuracy of 0.881 (dataset 1), 0.897 (dataset 2), and 0.807 (dataset 3) based on Dice similarity coefficient metric, offering significant performance improvements with a smaller atlas library (< 10) over compared methods. We also evaluate the diagnostic utility of AWoL-MRF by analyzing the volume differences per disease category in the ADNI1: Complete Screening dataset. We have made the source code for AWoL-MRF public at: https://github.com/CobraLab/AWoL-MRF.

MUSE: MUlti-atlas region Segmentation utilizing Ensembles of registration algorithms and parameters, and locally optimal atlas selection

Atlas-based automated anatomical labeling is a fundamental tool in medical image segmentation, as it defines regions of interest for subsequent analysis of structural and functional image data. The extensive investigation of multi-atlas warping and fusion techniques over the past 5 or more years has clearly demonstrated the advantages of consensus-based segmentation. However, the common approach is to use multiple atlases with a single registration method and parameter set, which is not necessarily optimal for every individual scan, anatomical region, and problem/data-type. Different registration criteria and parameter sets yield different solutions, each providing complementary information. Herein, we present a consensus labeling framework that generates a broad ensemble of labeled atlases in target image space via the use of several warping algorithms, regularization parameters, and atlases. The label fusion integrates two complementary sources of information: a local similarity ranking to select locally optimal atlases and a boundary modulation term to refine the segmentation consistently with the target image's intensity profile. The ensemble approach consistently outperforms segmentations using individual warping methods alone, achieving high accuracy on several benchmark datasets. The MUSE methodology has been used for processing thousands of scans from various datasets, producing robust and consistent results. MUSE is publicly available both as a downloadable software package, and as an application that can be run on the CBICA Image Processing Portal (https://ipp.cbica.upenn.edu), a web based platform for remote processing of medical images.

Geodesic Information Flows: Spatially-Variant Graphs and Their Application to Segmentation and Fusion

Clinical annotations, such as voxel-wise binary or probabilistic tissue segmentations, structural parcellations, pathological regions-of-interest and anatomical landmarks are key to many clinical studies. However, due to the time consuming nature of manually generating these annotations, they tend to be scarce and limited to small subsets of data. This work explores a novel framework to propagate voxel-wise annotations between morphologically dissimilar images by diffusing and mapping the available examples through intermediate steps. A spatially-variant graph structure connecting morphologically similar subjects is introduced over a database of images, enabling the gradual diffusion of information to all the subjects, even in the presence of large-scale morphological variability. We illustrate the utility of the proposed framework on two example applications: brain parcellation using categorical labels and tissue segmentation using probabilistic features. The application of the proposed method to categorical label fusion showed highly statistically significant improvements when compared to state-of-the-art methodologies. Significant improvements were also observed when applying the proposed framework to probabilistic tissue segmentation of both synthetic and real data, mainly in the presence of large morphological variability.

Multi-atlas segmentation of biomedical images: A survey

Multi-atlas segmentation (MAS), first introduced and popularized by the pioneering work of Rohlfing, et al. (2004), Klein, et al. (2005), and Heckemann, et al. (2006), is becoming one of the most widely-used and successful image segmentation techniques in biomedical applications. By manipulating and utilizing the entire dataset of "atlases" (training images that have been previously labeled, e.g., manually by an expert), rather than some model-based average representation, MAS has the flexibility to better capture anatomical variation, thus offering superior segmentation accuracy. This benefit, however, typically comes at a high computational cost. Recent advancements in computer hardware and image processing software have been instrumental in addressing this challenge and facilitated the wide adoption of MAS. Today, MAS has come a long way and the approach includes a wide array of sophisticated algorithms that employ ideas from machine learning, probabilistic modeling, optimization, and computer vision, among other fields. This paper presents a survey of published MAS algorithms and studies that have applied these methods to various biomedical problems. In writing this survey, we have three distinct aims. Our primary goal is to document how MAS was originally conceived, later evolved, and now relates to alternative methods. Second, this paper is intended to be a detailed reference of past research activity in MAS, which now spans over a decade (2003-2014) and entails novel methodological developments and application-specific solutions. Finally, our goal is to also present a perspective on the future of MAS, which, we believe, will be one of the dominant approaches in biomedical image segmentation.

Evaluation of Cross-Protocol Stability of a Fully Automated Brain Multi-Atlas Parcellation Tool

Brain parcellation tools based on multiple-atlas algorithms have recently emerged as a promising method with which to accurately define brain structures. When dealing with data from various sources, it is crucial that these tools are robust for many different imaging protocols. In this study, we tested the robustness of a multiple-atlas, likelihood fusion algorithm using Alzheimer’s Disease Neuroimaging Initiative (ADNI) data with six different protocols, comprising three manufacturers and two magnetic field strengths. The entire brain was parceled into five different levels of granularity. In each level, which defines a set of brain structures, ranging from eight to 286 regions, we evaluated the variability of brain volumes related to the protocol, age, and diagnosis (healthy or Alzheimer’s disease). Our results indicated that, with proper pre-processing steps, the impact of different protocols is minor compared to biological effects, such as age and pathology. A precise knowledge of the sources of data variation enables sufficient statistical power and ensures the reliability of an anatomical analysis when using this automated brain parcellation tool on datasets from various imaging protocols, such as clinical databases.

Local manifold learning for multiatlas segmentation: application to hippocampal segmentation in healthy population and Alzheimer's disease

AIMS

Automated hippocampal segmentation is an important issue in many neuroscience studies.

METHODS

We presented and evaluated a novel segmentation method that utilized a manifold learning technique under the multiatlas-based segmentation scenario. A manifold representation of local patches for each voxel was achieved by applying an Isomap algorithm, which can then be used to obtain spatially local weights of atlases for label fusion. The obtained atlas weights potentially depended on all pairwise similarities of the population, which is in contrast to most existing label fusion methods that only rely on similarities between the target image and the atlases. The performance of the proposed method was evaluated for hippocampal segmentation and compared with two representative local weighted label fusion methods, that is, local majority voting and local weighted inverse distance voting, on an in-house dataset of 28 healthy adolescents (age range: 10-17 years) and two ADNI datasets of 100 participants (age range: 60-89 years). We also implemented hippocampal volumetric analysis and evaluated segmentation performance using atlases from a different dataset.

RESULTS

The median Dice similarities obtained by our proposed method were approximately 0.90 for healthy subjects and above 0.88 for two mixed diagnostic groups of ADNI subjects.

CONCLUSION

The experimental results demonstrated that the proposed method could obtain consistent and significant improvements over label fusion strategies that are implemented in the original space.

Multi-atlas segmentation with joint label fusion and corrective learning-an open source implementation

Label fusion based multi-atlas segmentation has proven to be one of the most competitive techniques for medical image segmentation. This technique transfers segmentations from expert-labeled images, called atlases, to a novel image using deformable image registration. Errors produced by label transfer are further reduced by label fusion that combines the results produced by all atlases into a consensus solution. Among the proposed label fusion strategies, weighted voting with spatially varying weight distributions derived from atlas-target intensity similarity is a simple and highly effective label fusion technique. However, one limitation of most weighted voting methods is that the weights are computed independently for each atlas, without taking into account the fact that different atlases may produce similar label errors. To address this problem, we recently developed the joint label fusion technique and the corrective learning technique, which won the first place of the 2012 MICCAI Multi-Atlas Labeling Challenge and was one of the top performers in 2013 MICCAI Segmentation: Algorithms, Theory and Applications (SATA) challenge. To make our techniques more accessible to the scientific research community, we describe an Insight-Toolkit based open source implementation of our label fusion methods. Our implementation extends our methods to work with multi-modality imaging data and is more suitable for segmentation problems with multiple labels. We demonstrate the usage of our tools through applying them to the 2012 MICCAI Multi-Atlas Labeling Challenge brain image dataset and the 2013 SATA challenge canine leg image dataset. We report the best results on these two datasets so far.

Multi-atlas segmentation with joint label fusion and corrective learning-an open source implementation

Label fusion based multi-atlas segmentation has proven to be one of the most competitive techniques for medical image segmentation. This technique transfers segmentations from expert-labeled images, called atlases, to a novel image using deformable image registration. Errors produced by label transfer are further reduced by label fusion that combines the results produced by all atlases into a consensus solution. Among the proposed label fusion strategies, weighted voting with spatially varying weight distributions derived from atlas-target intensity similarity is a simple and highly effective label fusion technique. However, one limitation of most weighted voting methods is that the weights are computed independently for each atlas, without taking into account the fact that different atlases may produce similar label errors. To address this problem, we recently developed the joint label fusion technique and the corrective learning technique, which won the first place of the 2012 MICCAI Multi-Atlas Labeling Challenge and was one of the top performers in 2013 MICCAI Segmentation: Algorithms, Theory and Applications (SATA) challenge. To make our techniques more accessible to the scientific research community, we describe an Insight-Toolkit based open source implementation of our label fusion methods. Our implementation extends our methods to work with multi-modality imaging data and is more suitable for segmentation problems with multiple labels. We demonstrate the usage of our tools through applying them to the 2012 MICCAI Multi-Atlas Labeling Challenge brain image dataset and the 2013 SATA challenge canine leg image dataset. We report the best results on these two datasets so far.