A generative model for image segmentation based on label fusion

Simultaneous total intracranial volume and posterior fossa volume estimation using multi-atlas label fusion

Total intracranial volume (TICV) is an essential covariate in brain volumetric analyses. The prevalent brain imaging software packages provide automatic TICV estimates. FreeSurfer and FSL estimate TICV using a scaling factor while SPM12 accumulates probabilities of brain tissues. None of the three provide explicit skull/CSF boundary (SCB) since it is challenging to distinguish these dark structures in a T1-weighted image. However, explicit SCB not only leads to a natural way of obtaining TICV (i.e., counting voxels inside the skull) but also allows sub-definition of TICV, for example, the posterior fossa volume (PFV). In this article, they proposed to use multi-atlas label fusion to obtain TICV and PFV simultaneously. The main contributions are: (1) TICV and PFV are simultaneously obtained with explicit SCB from a single T1-weighted image. (2) TICV and PFV labels are added to the widely used BrainCOLOR atlases. (3) Detailed mathematical derivation of non-local spatial STAPLE (NLSS) label fusion is presented. As the skull is clearly distinguished in CT images, we use a semi-manual procedure to obtain atlases with TICV and PFV labels using 20 subjects who both have a MR and CT scan. The proposed method provides simultaneous TICV and PFV estimation while achieving more accurate TICV estimation compared with FreeSurfer, FSL, SPM12, and the previously proposed STAPLE based approach. The newly developed TICV and PFV labels for the OASIS BrainCOLOR atlases provide acceptable performance, which enables simultaneous TICV and PFV estimation during whole brain segmentation. The NLSS method and the new atlases have been made freely available. Hum Brain Mapp 38:599-616, 2017. © 2016 Wiley Periodicals, Inc.

Direct estimation of patient attributes from anatomical MRI based on multi-atlas voting

MRI brain atlases are widely used for automated image segmentation, and in particular, recent developments in multi-atlas techniques have shown highly accurate segmentation results. In this study, we extended the role of the atlas library from mere anatomical reference to a comprehensive knowledge database with various patient attributes, such as demographic, functional, and diagnostic information. In addition to using the selected (heavily-weighted) atlases to achieve high segmentation accuracy, we tested whether the non-anatomical attributes of the selected atlases could be used to estimate patient attributes. This can be considered a context-based image retrieval (CBIR) approach, embedded in the multi-atlas framework. We first developed an image similarity measurement to weigh the atlases on a structure-by-structure basis, and then, the attributes of the multiple atlases were weighted to estimate the patient attributes. We tested this concept first by estimating age in a normal population; we then performed functional and diagnostic estimations in Alzheimer's disease patients. The accuracy of the estimated patient attributes was measured against the actual clinical data, and the performance was compared to conventional volumetric analysis. The proposed CBIR framework by multi-atlas voting would be the first step toward a knowledge-based support system for quantitative radiological image reading and diagnosis.

•

Patient attributes are estimated directly by retrieving information from multi-atlas.

•

Non-imaging attributes of the atlases are weighted to estimate patient attributes.

•

The method achieved high accuracy in estimating age in the normal population.

•

The method can estimate functional and diagnostic attributes in dementia patients.

•

The estimation accuracy was higher than volumetric analysis in subcortical areas.

Fast and sequence-adaptive whole-brain segmentation using parametric Bayesian modeling

Quantitative analysis of magnetic resonance imaging (MRI) scans of the brain requires accurate automated segmentation of anatomical structures. A desirable feature for such segmentation methods is to be robust against changes in acquisition platform and imaging protocol. In this paper we validate the performance of a segmentation algorithm designed to meet these requirements, building upon generative parametric models previously used in tissue classification. The method is tested on four different datasets acquired with different scanners, field strengths and pulse sequences, demonstrating comparable accuracy to state-of-the-art methods on T1-weighted scans while being one to two orders of magnitude faster. The proposed algorithm is also shown to be robust against small training datasets, and readily handles images with different MRI contrast as well as multi-contrast data.

Brain tumor segmentation from multimodal magnetic resonance images via sparse representation

OBJECTIVE

Accurately segmenting and quantifying brain gliomas from magnetic resonance (MR) images remains a challenging task because of the large spatial and structural variability among brain tumors. To develop a fully automatic and accurate brain tumor segmentation algorithm, we present a probabilistic model of multimodal MR brain tumor segmentation. This model combines sparse representation and the Markov random field (MRF) to solve the spatial and structural variability problem.

METHODS

We formulate the tumor segmentation problem as a multi-classification task by labeling each voxel as the maximum posterior probability. We estimate the maximum a posteriori (MAP) probability by introducing the sparse representation into a likelihood probability and a MRF into the prior probability. Considering the MAP as an NP-hard problem, we convert the maximum posterior probability estimation into a minimum energy optimization problem and employ graph cuts to find the solution to the MAP estimation.

RESULTS

Our method is evaluated using the Brain Tumor Segmentation Challenge 2013 database (BRATS 2013) and obtained Dice coefficient metric values of 0.85, 0.75, and 0.69 on the high-grade Challenge data set, 0.73, 0.56, and 0.54 on the high-grade Challenge LeaderBoard data set, and 0.84, 0.54, and 0.57 on the low-grade Challenge data set for the complete, core, and enhancing regions.

CONCLUSIONS

The experimental results show that the proposed algorithm is valid and ranks 2nd compared with the state-of-the-art tumor segmentation algorithms in the MICCAI BRATS 2013 challenge.

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.

The feasibility of atlas-based automatic segmentation of MRI for H&N radiotherapy planning

Atlas‐based autosegmentation is an established tool for segmenting structures for CT‐planned head and neck radiotherapy. MRI is being increasingly integrated into the planning process. The aim of this study is to assess the feasibility of MRI‐based, atlas‐based autosegmentation for organs at risk (OAR) and lymph node levels, and to compare the segmentation accuracy with CT‐based autosegmentation. Fourteen patients with locally advanced head and neck cancer in a prospective imaging study underwent a T1‐weighted MRI and a PET‐CT (with dedicated contrast‐enhanced CT) in an immobilization mask. Organs at risk (orbits, parotids, brainstem, and spinal cord) and the left level II lymph node region were manually delineated on the CT and MRI separately. A ‘leave one out’ approach was used to automatically segment structures onto the remaining images separately for CT and MRI. Contour comparison was performed using multiple positional metrics: Dice index, mean distance to conformity (MDC), sensitivity index (Se Idx), and inclusion index (Incl Idx). Automatic segmentation using MRI of orbits, parotids, brainstem, and lymph node level was acceptable with a DICE coefficient of 0.73−0.91, MDC 2.0−5.1 mm, Se Idx 0.64−0.93, Incl Idx 0.76−0.93. Segmentation of the spinal cord was poor (Dice coefficient 0.37). The process of automatic segmentation was significantly better on MRI compared to CT for orbits, parotid glands, brainstem, and left lymph node level II by multiple positional metrics; spinal cord segmentation based on MRI was inferior compared with CT. Accurate atlas‐based automatic segmentation of OAR and lymph node levels is feasible using T1‐MRI; segmentation of the spinal cord was found to be poor. Comparison with CT‐based automatic segmentation suggests that the process is equally as, or more accurate, using MRI. These results support further translation of MRI‐based segmentation methodology into clinical practice.

PACS number(s): 87.55.de

Investigation of Bias in Continuous Medical Image Label Fusion

Image labeling is essential for analyzing morphometric features in medical imaging data. Labels can be obtained by either human interaction or automated segmentation algorithms, both of which suffer from errors. The Simultaneous Truth and Performance Level Estimation (STAPLE) algorithm for both discrete-valued and continuous-valued labels has been proposed to find the consensus fusion while simultaneously estimating rater performance. In this paper, we first show that the previously reported continuous STAPLE in which bias and variance are used to represent rater performance yields a maximum likelihood solution in which bias is indeterminate. We then analyze the major cause of the deficiency and evaluate two classes of auxiliary bias estimation processes, one that estimates the bias as part of the algorithm initialization and the other that uses a maximum a posteriori criterion with a priori probabilities on the rater bias. We compare the efficacy of six methods, three variants from each class, in simulations and through empirical human rater experiments. We comment on their properties, identify deficient methods, and propose effective methods as solution.

Evaluation of Six Registration Methods for the Human Abdomen on Clinically Acquired CT

OBJECTIVE

This work evaluates current 3-D image registration tools on clinically acquired abdominal computed tomography (CT) scans.

METHODS

Thirteen abdominal organs were manually labeled on a set of 100 CT images, and the 100 labeled images (i.e., atlases) were pairwise registered based on intensity information with six registration tools (FSL, ANTS-CC, ANTS-QUICK-MI, IRTK, NIFTYREG, and DEEDS). The Dice similarity coefficient (DSC), mean surface distance, and Hausdorff distance were calculated on the registered organs individually. Permutation tests and indifference-zone ranking were performed to examine the statistical and practical significance, respectively.

RESULTS

The results suggest that DEEDS yielded the best registration performance. However, due to the overall low DSC values, and substantial portion of low-performing outliers, great care must be taken when image registration is used for local interpretation of abdominal CT.

CONCLUSION

There is substantial room for improvement in image registration for abdominal CT.

SIGNIFICANCE

All data and source code are available so that innovations in registration can be directly compared with the current generation of tools without excessive duplication of effort.

An Empirical Study Into Annotator Agreement, Ground Truth Estimation, and Algorithm Evaluation

Although agreement between the annotators who mark feature locations within images has been studied in the past from a statistical viewpoint, little work has attempted to quantify the extent to which this phenomenon affects the evaluation of foreground-background segmentation algorithms. Many researchers utilize ground truth (GT) in experimentation and more often than not this GT is derived from one annotator's opinion. How does the difference in opinion affects an algorithm's evaluation? A methodology is applied to four image-processing problems to quantify the interannotator variance and to offer insight into the mechanisms behind agreement and the use of GT. It is found that when detecting linear structures, annotator agreement is very low. The agreement in a structure's position can be partially explained through basic image properties. Automatic segmentation algorithms are compared with annotator agreement and it is found that there is a clear relation between the two. Several GT estimation methods are used to infer a number of algorithm performances. It is found that the rank of a detector is highly dependent upon the method used to form the GT, and that although STAPLE and LSML appear to represent the mean of the performance measured using individual annotations, when there are few annotations, or there is a large variance in them, these estimates tend to degrade. Furthermore, one of the most commonly adopted combination methods-consensus voting-accentuates more obvious features, resulting in an overestimation of performance. It is concluded that in some data sets, it is not possible to confidently infer an algorithm ranking when evaluating upon one GT.

Consistent cortical reconstruction and multi-atlas brain segmentation

Whole brain segmentation and cortical surface reconstruction are two essential techniques for investigating the human brain. Spatial inconsistences, which can hinder further integrated analyses of brain structure, can result due to these two tasks typically being conducted independently of each other. FreeSurfer obtains self-consistent whole brain segmentations and cortical surfaces. It starts with subcortical segmentation, then carries out cortical surface reconstruction, and ends with cortical segmentation and labeling. However, this "segmentation to surface to parcellation" strategy has shown limitations in various cohorts such as older populations with large ventricles. In this work, we propose a novel "multi-atlas segmentation to surface" method called Multi-atlas CRUISE (MaCRUISE), which achieves self-consistent whole brain segmentations and cortical surfaces by combining multi-atlas segmentation with the cortical reconstruction method CRUISE. A modification called MaCRUISE(+) is designed to perform well when white matter lesions are present. Comparing to the benchmarks CRUISE and FreeSurfer, the surface accuracy of MaCRUISE and MaCRUISE(+) is validated using two independent datasets with expertly placed cortical landmarks. A third independent dataset with expertly delineated volumetric labels is employed to compare segmentation performance. Finally, 200MR volumetric images from an older adult sample are used to assess the robustness of MaCRUISE and FreeSurfer. The advantages of MaCRUISE are: (1) MaCRUISE constructs self-consistent voxelwise segmentations and cortical surfaces, while MaCRUISE(+) is robust to white matter pathology. (2) MaCRUISE achieves more accurate whole brain segmentations than independently conducting the multi-atlas segmentation. (3) MaCRUISE is comparable in accuracy to FreeSurfer (when FreeSurfer does not exhibit global failures) while achieving greater robustness across an older adult population. MaCRUISE has been made freely available in open source.

Sampling image segmentations for uncertainty quantification

In this paper, we introduce a method to automatically produce plausible image segmentation samples from a single expert segmentation. A probability distribution of image segmentation boundaries is defined as a Gaussian process, which leads to segmentations which are spatially coherent and consistent with the presence of salient borders in the image. The proposed approach is computationally efficient, and generates visually plausible samples. The variability between the samples is mainly governed by a parameter which may be correlated with a simple Dice's coefficient, or easily set by the user from the definition of probable regions of interest. The method is extended to the case of several neighboring structures, but also to account for under or over segmentation, and the presence of excluded regions. We also detail a method to sample segmentations with more general non-stationary covariance functions which relies on supervoxels. Furthermore, we compare the generated segmentation samples with several manual clinical segmentations of a brain tumor. Finally, we show how this approach can have useful applications in the field of uncertainty quantification, and an illustration is provided in radiotherapy planning, where segmentation sampling is applied to both the clinical target volume and the organs at risk.

Atlas-based rib-bone detection in chest X-rays

This paper investigates using rib-bone atlases for automatic detection of rib-bones in chest X-rays (CXRs). We built a system that takes patient X-ray and model atlases as input and automatically computes the posterior rib borders with high accuracy and efficiency. In addition to conventional atlas, we propose two alternative atlases: (i) automatically computed rib bone models using Computed Tomography (CT) scans, and (ii) dual energy CXRs. We test the proposed approach with each model on 25 CXRs from the Japanese Society of Radiological Technology (JSRT) dataset and another 25 CXRs from the National Library of Medicine CXR dataset. We achieve an area under the ROC curve (AUC) of about 95% for Montgomery and 91% for JSRT datasets. Using the optimal operating point of the ROC curve, we achieve a segmentation accuracy of 88.91±1.8% for Montgomery and 85.48±3.3% for JSRT datasets. Our method produces comparable results with the state-of-the-art algorithms. The performance of our method is also excellent on challenging X-rays as it successfully addressed the rib-shape variance between patients and number of visible rib-bones due to patient respiration.

A Patch-Based Approach for the Segmentation of Pathologies: Application to Glioma Labelling

In this paper, we describe a novel and generic approach to address fully-automatic segmentation of brain tumors by using multi-atlas patch-based voting techniques. In addition to avoiding the local search window assumption, the conventional patch-based framework is enhanced through several simple procedures: an improvement of the training dataset in terms of both label purity and intensity statistics, augmented features to implicitly guide the nearest-neighbor-search, multi-scale patches, invariance to cube isometries, stratification of the votes with respect to cases and labels. A probabilistic model automatically delineates regions of interest enclosing high-probability tumor volumes, which allows the algorithm to achieve highly competitive running time despite minimal processing power and resources. This method was evaluated on Multimodal Brain Tumor Image Segmentation challenge datasets. State-of-the-art results are achieved, with a limited learning stage thus restricting the risk of overfit. Moreover, segmentation smoothness does not involve any post-processing.

Deformable MR Prostate Segmentation via Deep Feature Learning and Sparse Patch Matching

Automatic and reliable segmentation of the prostate is an important but difficult task for various clinical applications such as prostate cancer radiotherapy. The main challenges for accurate MR prostate localization lie in two aspects: (1) inhomogeneous and inconsistent appearance around prostate boundary, and (2) the large shape variation across different patients. To tackle these two problems, we propose a new deformable MR prostate segmentation method by unifying deep feature learning with the sparse patch matching. First, instead of directly using handcrafted features, we propose to learn the latent feature representation from prostate MR images by the stacked sparse auto-encoder (SSAE). Since the deep learning algorithm learns the feature hierarchy from the data, the learned features are often more concise and effective than the handcrafted features in describing the underlying data. To improve the discriminability of learned features, we further refine the feature representation in a supervised fashion. Second, based on the learned features, a sparse patch matching method is proposed to infer a prostate likelihood map by transferring the prostate labels from multiple atlases to the new prostate MR image. Finally, a deformable segmentation is used to integrate a sparse shape model with the prostate likelihood map for achieving the final segmentation. The proposed method has been extensively evaluated on the dataset that contains 66 T2-wighted prostate MR images. Experimental results show that the deep-learned features are more effective than the handcrafted features in guiding MR prostate segmentation. Moreover, our method shows superior performance than other state-of-the-art segmentation methods.

A review of heart chamber segmentation for structural and functional analysis using cardiac magnetic resonance imaging

Cardiovascular magnetic resonance (CMR) has become a key imaging modality in clinical cardiology practice due to its unique capabilities for non-invasive imaging of the cardiac chambers and great vessels. A wide range of CMR sequences have been developed to assess various aspects of cardiac structure and function, and significant advances have also been made in terms of imaging quality and acquisition times. A lot of research has been dedicated to the development of global and regional quantitative CMR indices that help the distinction between health and pathology. The goal of this review paper is to discuss the structural and functional CMR indices that have been proposed thus far for clinical assessment of the cardiac chambers. We include indices definitions, the requirements for the calculations, exemplar applications in cardiovascular diseases, and the corresponding normal ranges. Furthermore, we review the most recent state-of-the art techniques for the automatic segmentation of the cardiac boundaries, which are necessary for the calculation of the CMR indices. Finally, we provide a detailed discussion of the existing literature and of the future challenges that need to be addressed to enable a more robust and comprehensive assessment of the cardiac chambers in clinical practice.

Automatic labeling of MR brain images by hierarchical learning of atlas forests

PURPOSE

Automatic brain image labeling is highly demanded in the field of medical image analysis. Multiatlas-based approaches are widely used due to their simplicity and robustness in applications. Also, random forest technique is recognized as an efficient method for labeling, although there are several existing limitations. In this paper, the authors intend to address those limitations by proposing a novel framework based on the hierarchical learning of atlas forests.

METHODS

Their proposed framework aims to train a hierarchy of forests to better correlate voxels in the MR images with their corresponding labels. There are two specific novel strategies for improving brain image labeling. First, different from the conventional ways of using a single level of random forests for brain labeling, the authors design a hierarchical structure to incorporate multiple levels of forests. In particular, each atlas forest in the bottom level is trained in accordance with each individual atlas, and then the bottom-level forests are clustered based on their capabilities in labeling. For each clustered group, the authors retrain a new representative forest in the higher level by using all atlases associated with the lower-level atlas forests in the current group, as well as the tentative label maps yielded from the lower level. This clustering and retraining procedure is conducted iteratively to yield a hierarchical structure of forests. Second, in the testing stage, the authors also present a novel atlas forest selection method to determine an optimal set of atlas forests from the constructed hierarchical structure (by disabling those nonoptimal forests) for accurately labeling the test image.

RESULTS

For validating their proposed framework, the authors evaluate it on the public datasets, including Alzheimer's disease neuroimaging initiative, Internet brain segmentation repository, and LONI LPBA40. The authors compare the results with the conventional approaches. The experiments show that the use of the two novel strategies can significantly improve the labeling performance. Note that when more levels are constructed in the hierarchy, the labeling performance can be further improved, but more computational time will be also required.

CONCLUSIONS

The authors have proposed a novel multiatlas-based framework for automatic and accurate labeling of brain anatomies, which can achieve accurate labeling results for MR brain images.

Improving Cerebellar Segmentation with Statistical Fusion

The cerebellum is a somatotopically organized central component of the central nervous system well known to be involved with motor coordination and increasingly recognized roles in cognition and planning. Recent work in multi-atlas labeling has created methods that offer the potential for fully automated 3-D parcellation of the cerebellar lobules and vermis (which are organizationally equivalent to cortical gray matter areas). This work explores the trade offs of using different statistical fusion techniques and post hoc optimizations in two datasets with distinct imaging protocols. We offer a novel fusion technique by extending the ideas of the Selective and Iterative Method for Performance Level Estimation (SIMPLE) to a patch-based performance model. We demonstrate the effectiveness of our algorithm, Non-Local SIMPLE, for segmentation of a mixed population of healthy subjects and patients with severe cerebellar anatomy. Under the first imaging protocol, we show that Non-Local SIMPLE outperforms previous gold-standard segmentation techniques. In the second imaging protocol, we show that Non-Local SIMPLE outperforms previous gold standard techniques but is outperformed by a non-locally weighted vote with the deeper population of atlases available. This work advances the state of the art in open source cerebellar segmentation algorithms and offers the opportunity for routinely including cerebellar segmentation in magnetic resonance imaging studies that acquire whole brain T1-weighted volumes with approximately 1 mm isotropic resolution.

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.

Patch-based augmentation of Expectation-Maximization for brain MRI tissue segmentation at arbitrary age after premature birth

Accurate automated tissue segmentation of premature neonatal magnetic resonance images is a crucial task for quantification of brain injury and its impact on early postnatal growth and later cognitive development. In such studies it is common for scans to be acquired shortly after birth or later during the hospital stay and therefore occur at arbitrary gestational ages during a period of rapid developmental change. It is important to be able to segment any of these scans with comparable accuracy. Previous work on brain tissue segmentation in premature neonates has focused on segmentation at specific ages. Here we look at solving the more general problem using adaptations of age specific atlas based methods and evaluate this using a unique manually traced database of high resolution images spanning 20 gestational weeks of development. We examine the complimentary strengths of age specific atlas-based Expectation-Maximization approaches and patch-based methods for this problem and explore the development of two new hybrid techniques, patch-based augmentation of Expectation-Maximization with weighted fusion and a spatial variability constrained patch search. The former approach seeks to combine the advantages of both atlas- and patch-based methods by learning from the performance of the two techniques across the brain anatomy at different developmental ages, while the latter technique aims to use anatomical variability maps learnt from atlas training data to locally constrain the patch-based search range. The proposed approaches were evaluated using leave-one-out cross-validation. Compared with the conventional age specific atlas-based segmentation and direct patch based segmentation, both new approaches demonstrate improved accuracy in the automated labeling of cortical gray matter, white matter, ventricles and sulcal cortical-spinal fluid regions, while maintaining comparable results in deep gray matter.

Multi-atlas label fusion using hybrid of discriminative and generative classifiers for segmentation of cardiac MR images

Multi-atlas segmentation first registers each atlas image to the target image and transfers the label of atlas image to the coordinate system of the target image. The transferred labels are then combined, using a label fusion algorithm. In this paper, we propose a novel label fusion method which aggregates discriminative learning and generative modeling for segmentation of cardiac MR images. First, a probabilistic Random Forest classifier is trained as a discriminative model to obtain the prior probability of a label at the given voxel of the target image. Then, a probability distribution of image patches is modeled using Gaussian Mixture Model for each label, providing the likelihood of the voxel belonging to the label. The final label posterior is obtained by combining the classification score and the likelihood score under Bayesian rule. Comparative study performed on MICCAI 2013 SATA Segmentation Challenge demonstrates that our proposed hybrid label fusion algorithm is accurate than other five state-of-the-art label fusion methods. The proposed method obtains dice similarity coefficient of 0.94 and 0.92 in segmenting epicardium and endocardium respectively. Moreover, our label fusion method achieves more accurate segmentation results compared to four other label fusion methods.

Volumetric and fiber-tracing MRI methods for gray and white matter

Magnetic resonance imaging (MRI) is capable of generating high-resolution brain images with fine anatomic detail and unique tissue contrasts that reveal structures that are not visible to the eye. Sharply defined gray- and white-matter interfaces allow for quantitative anatomic analysis that can be accurately performed with largely automated segmentation methods. In an analogous fashion, diffusion MRI in the brain provides structural information based on contrasts derived from the diffusivity of water in brain tissue, which can highlight the orientation of neuronal axons. Also using largely automated methods, diffusion MRI can be used to generate models of white-matter tracts throughout the brain, a method known as tractography, as well as characterize the microstructural integrity of neuronal axons. Tractographic analysis has helped to define connectivity in the brain that powerfully informs understanding of brain function, and, together with other diffusion metrics, is useful in evaluation of the normal and diseased brain. The quantitative methods of brain segmentation, tractography, and diffusion MRI extend MRI into a realm beyond visual inspection and provide otherwise unachievable sensitivity and specificity in the analysis of brain structure and function.

Automated segmentation of dental CBCT image with prior-guided sequential random forests

PURPOSE

Cone-beam computed tomography (CBCT) is an increasingly utilized imaging modality for the diagnosis and treatment planning of the patients with craniomaxillofacial (CMF) deformities. Accurate segmentation of CBCT image is an essential step to generate 3D models for the diagnosis and treatment planning of the patients with CMF deformities. However, due to the image artifacts caused by beam hardening, imaging noise, inhomogeneity, truncation, and maximal intercuspation, it is difficult to segment the CBCT.

METHODS

In this paper, the authors present a new automatic segmentation method to address these problems. Specifically, the authors first employ a majority voting method to estimate the initial segmentation probability maps of both mandible and maxilla based on multiple aligned expert-segmented CBCT images. These probability maps provide an important prior guidance for CBCT segmentation. The authors then extract both the appearance features from CBCTs and the context features from the initial probability maps to train the first-layer of random forest classifier that can select discriminative features for segmentation. Based on the first-layer of trained classifier, the probability maps are updated, which will be employed to further train the next layer of random forest classifier. By iteratively training the subsequent random forest classifier using both the original CBCT features and the updated segmentation probability maps, a sequence of classifiers can be derived for accurate segmentation of CBCT images.

RESULTS

Segmentation results on CBCTs of 30 subjects were both quantitatively and qualitatively validated based on manually labeled ground truth. The average Dice ratios of mandible and maxilla by the authors' method were 0.94 and 0.91, respectively, which are significantly better than the state-of-the-art method based on sparse representation (p-value < 0.001).

CONCLUSIONS

The authors have developed and validated a novel fully automated method for CBCT segmentation.

Improved segmentation of cerebellar structures in children

BACKGROUND

Consistent localization of cerebellar cortex in a standard coordinate system is important for functional studies and detection of anatomical alterations in studies of morphometry. To date, no pediatric cerebellar atlas is available.

NEW METHOD

The probabilistic Cape Town Pediatric Cerebellar Atlas (CAPCA18) was constructed in the age-appropriate National Institute of Health Pediatric Database asymmetric template space using manual tracings of 16 cerebellar compartments in 18 healthy children (9-13 years) from Cape Town, South Africa. The individual atlases of the training subjects were also used to implement multi atlas label fusion using multi atlas majority voting (MAMV) and multi atlas generative model (MAGM) approaches. Segmentation accuracy in 14 test subjects was compared for each method to 'gold standard' manual tracings.

RESULTS

Spatial overlap between manual tracings and CAPCA18 automated segmentation was 73% or higher for all lobules in both hemispheres, except VIIb and X. Automated segmentation using MAGM yielded the best segmentation accuracy over all lobules (mean Dice Similarity Coefficient 0.76; range 0.55-0.91; mean Hausdorff distance 0.9 mm; range 0.8-2.7 mm).

COMPARISON WITH EXISTING METHODS

In all lobules, spatial overlap of CAPCA18 segmentations with manual tracings was similar or higher than those obtained with SUIT (spatially unbiased infra-tentorial template), providing additional evidence of the benefits of an age appropriate atlas. MAGM segmentation accuracy was comparable to values reported recently by Park et al. (Neuroimage 2014;95(1):217) in adults (across all lobules mean DSC=0.73, range 0.40-0.89).

CONCLUSIONS

CAPCA18 and the associated multi-subject atlases of the training subjects yield improved segmentation of cerebellar structures in children.

Contour-Driven Atlas-Based Segmentation

We propose new methods for automatic segmentation of images based on an atlas of manually labeled scans and contours in the image. First, we introduce a Bayesian framework for creating initial label maps from manually annotated training images. Within this framework, we model various registration- and patch-based segmentation techniques by changing the deformation field prior. Second, we perform contour-driven regression on the created label maps to refine the segmentation. Image contours and image parcellations give rise to non-stationary kernel functions that model the relationship between image locations. Setting the kernel to the covariance function in a Gaussian process establishes a distribution over label maps supported by image structures. Maximum a posteriori estimation of the distribution over label maps conditioned on the outcome of the atlas-based segmentation yields the refined segmentation. We evaluate the segmentation in two clinical applications: the segmentation of parotid glands in head and neck CT scans and the segmentation of the left atrium in cardiac MR angiography images.

Automatic segmentation of mandible in panoramic x-ray

As the panoramic x-ray is the most common extraoral radiography in dentistry, segmentation of its anatomical structures facilitates diagnosis and registration of dental records. This study presents a fast and accurate method for automatic segmentation of mandible in panoramic x-rays. In the proposed four-step algorithm, a superior border is extracted through horizontal integral projections. A modified Canny edge detector accompanied by morphological operators extracts the inferior border of the mandible body. The exterior borders of ramuses are extracted through a contour tracing method based on the average model of mandible. The best-matched template is fetched from the atlas of mandibles to complete the contour of left and right processes. The algorithm was tested on a set of 95 panoramic x-rays. Evaluating the results against manual segmentations of three expert dentists showed that the method is robust. It achieved an average performance of [Formula: see text] in Dice similarity, specificity, and sensitivity.

Brain extraction in pediatric ADC maps, toward characterizing neuro-development in multi-platform and multi-institution clinical images

Apparent Diffusion Coefficient (ADC) maps can be used to characterize myelination and to detect abnormalities in the developing brain. However, given the normal variation in regional ADC with myelination, detection of abnormalities is difficult when based on visual assessment. Quantitative and automated analysis of pediatric ADC maps is thus desired but requires accurate brain extraction as the first step. Currently, most existing brain extraction methods are optimized for structural T1-weighted MR images of fully myelinated brains. Due to differences in age and image contrast, these approaches do not translate well to pediatric ADC maps. To address this problem, we present a multi-atlas brain extraction framework that has 1) specificity: designed and optimized specifically for pediatric ADC maps; 2) generality: applicable to multi-platform and multi-institution data, and to subjects at various neuro-developmental stages across the first 6 years of life; 3) accuracy: highly accurate compared to expert annotations; and 4) consistency: consistently accurate regardless of sources of data and ages of subjects. We show how we achieve these goals, via optimizing major components in a multi-atlas brain extraction framework, and via developing and evaluating new criteria for its atlas ranking component. Moreover, we demonstrate that these goals can be achieved with a fixed set of atlases and a fixed set of parameters, which opens doors for our optimized framework to be used in large-scale and multi-institution neuro-developmental and clinical studies. In a pilot study, we use this framework in a dataset containing scanner-generated ADC maps from 308 pediatric patients collected during the course of routine clinical care. Our framework leads to successful quantifications of the changes in whole-brain volumes and mean ADC values across the first 6 years of life.

Automated retinofugal visual pathway reconstruction with multi-shell HARDI and FOD-based analysis

Diffusion MRI tractography provides a non-invasive modality to examine the human retinofugal projection, which consists of the optic nerves, optic chiasm, optic tracts, the lateral geniculate nuclei (LGN) and the optic radiations. However, the pathway has several anatomic features that make it particularly challenging to study with tractography, including its location near blood vessels and bone-air interface at the base of the cerebrum, crossing fibers at the chiasm, somewhat-tortuous course around the temporal horn via Meyer's Loop, and multiple closely neighboring fiber bundles. To date, these unique complexities of the visual pathway have impeded the development of a robust and automated reconstruction method using tractography. To overcome these challenges, we develop a novel, fully automated system to reconstruct the retinofugal visual pathway from high-resolution diffusion imaging data. Using multi-shell, high angular resolution diffusion imaging (HARDI) data, we reconstruct precise fiber orientation distributions (FODs) with high order spherical harmonics (SPHARM) to resolve fiber crossings, which allows the tractography algorithm to successfully navigate the complicated anatomy surrounding the retinofugal pathway. We also develop automated algorithms for the identification of ROIs used for fiber bundle reconstruction. In particular, we develop a novel approach to extract the LGN region of interest (ROI) based on intrinsic shape analysis of a fiber bundle computed from a seed region at the optic chiasm to a target at the primary visual cortex. By combining automatically identified ROIs and FOD-based tractography, we obtain a fully automated system to compute the main components of the retinofugal pathway, including the optic tract and the optic radiation. We apply our method to the multi-shell HARDI data of 215 subjects from the Human Connectome Project (HCP). Through comparisons with post-mortem dissection measurements, we demonstrate the retinotopic organization of the optic radiation including a successful reconstruction of Meyer's loop. Then, using the reconstructed optic radiation bundle from the HCP cohort, we construct a probabilistic atlas and demonstrate its consistency with a post-mortem atlas. Finally, we generate a shape-based representation of the optic radiation for morphometry analysis.

Evaluation and optimization of the parameters used in multiple-atlas-based segmentation of prostate cancers in radiation therapy

OBJECTIVE

To evaluate and optimize the parameters used in multiple-atlas-based segmentation of prostate cancers in radiation therapy.

METHODS

A retrospective study was conducted, and the accuracy of the multiple-atlas-based segmentation was tested on 30 patients. The effect of library size (LS), number of atlases used for contour averaging and the contour averaging strategy were also studied. The autogenerated contours were compared with the manually drawn contours. Dice similarity coefficient (DSC) and Hausdorff distance were used to evaluate the segmentation agreement.

RESULTS

Mixed results were found between simultaneous truth and performance level estimation (STAPLE) and majority vote (MV) strategies. Multiple-atlas approaches were relatively insensitive to LS. A LS of ten was adequate, and further increase in the LS only showed insignificant gain. Multiple atlas performed better than single atlas for most of the time. Using more atlases did not guarantee better performance, with five atlases performing better than ten atlases. With our recommended setting, the median DSC for the bladder, rectum, prostate, seminal vesicle and femurs was 0.90, 0.77, 0.84, 0.56 and 0.95, respectively.

CONCLUSION

Our study shows that multiple-atlas-based strategies have better accuracy than single-atlas approach. STAPLE is preferred, and a LS of ten is adequate for prostate cases. Using five atlases for contour averaging is recommended. The contouring accuracy of seminal vesicle still needs improvement, and manual editing is still required for the other structures.

ADVANCES IN KNOWLEDGE

This article provides a better understanding of the influence of the parameters used in multiple-atlas-based segmentation of prostate cancers.

Resource atlases for multi-atlas brain segmentations with multiple ontology levels based on T1-weighted MRI

Technologies for multi-atlas brain segmentation of T1-weighted MRI images have rapidly progressed in recent years, with highly promising results. This approach, however, relies on a large number of atlases with accurate and consistent structural identifications. Here, we introduce our atlas inventories (n=90), which cover ages 4-82years with unique hierarchical structural definitions (286 structures at the finest level). This multi-atlas library resource provides the flexibility to choose appropriate atlases for various studies with different age ranges and structure-definition criteria. In this paper, we describe the details of the atlas resources and demonstrate the improved accuracy achievable with a dynamic age-matching approach, in which atlases that most closely match the subject's age are dynamically selected. The advanced atlas creation strategy, together with atlas pre-selection principles, is expected to support the further development of multi-atlas image segmentation.

Predicting Activation Across Individuals with Resting-State Functional Connectivity Based Multi-Atlas Label Fusion

The alignment of brain imaging data for functional neuroimaging studies is challenging due to the discrepancy between correspondence of morphology, and equivalence of functional role. In this paper we map functional activation areas across individuals by a multi-atlas label fusion algorithm in a functional space. We learn the manifold of resting-state fMRI signals in each individual, and perform manifold alignment in an embedding space. We then transfer activation predictions from a source population to a target subject via multi-atlas label fusion. The cost function is derived from the aligned manifolds, so that the resulting correspondences are derived based on the similarity of intrinsic connectivity architecture. Experiments show that the resulting label fusion predicts activation evoked by various experiment conditions with higher accuracy than relying on morphological alignment. Interestingly, the distribution of this gain is distributed heterogeneously across the cortex, and across tasks. This offers insights into the relationship between intrinsic connectivity, morphology and task activation. Practically, the mechanism can serve as prior, and provides an avenue to infer task-related activation in individuals for whom only resting data is available.

A Latent Source Model for Patch-Based Image Segmentation

Despite the popularity and empirical success of patch-based nearest-neighbor and weighted majority voting approaches to medical image segmentation, there has been no theoretical development on when, why, and how well these nonparametric methods work. We bridge this gap by providing a theoretical performance guarantee for nearest-neighbor and weighted majority voting segmentation under a new probabilistic model for patch-based image segmentation. Our analysis relies on a new local property for how similar nearby patches are, and fuses existing lines of work on modeling natural imagery patches and theory for nonparametric classification. We use the model to derive a new patch-based segmentation algorithm that iterates between inferring local label patches and merging these local segmentations to produce a globally consistent image segmentation. Many existing patch-based algorithms arise as special cases of the new algorithm.

Subject-Specific Sparse Dictionary Learning for Atlas-Based Brain MRI Segmentation

Quantitative measurements from segmentations of human brain magnetic resonance (MR) images provide important biomarkers for normal aging and disease progression. In this paper, we propose a patch-based tissue classification method from MR images that uses a sparse dictionary learning approach and atlas priors. Training data for the method consists of an atlas MR image, prior information maps depicting where different tissues are expected to be located, and a hard segmentation. Unlike most atlas-based classification methods that require deformable registration of the atlas priors to the subject, only affine registration is required between the subject and training atlas. A subject-specific patch dictionary is created by learning relevant patches from the atlas. Then the subject patches are modeled as sparse combinations of learned atlas patches leading to tissue memberships at each voxel. The combination of prior information in an example-based framework enables us to distinguish tissues having similar intensities but different spatial locations. We demonstrate the efficacy of the approach on the application of whole-brain tissue segmentation in subjects with healthy anatomy and normal pressure hydrocephalus, as well as lesion segmentation in multiple sclerosis patients. For each application, quantitative comparisons are made against publicly available state-of-the art approaches.

Multi-atlas learner fusion: An efficient segmentation approach for large-scale data

We propose multi-atlas learner fusion (MLF), a framework for rapidly and accurately replicating the highly accurate, yet computationally expensive, multi-atlas segmentation framework based on fusing local learners. In the largest whole-brain multi-atlas study yet reported, multi-atlas segmentations are estimated for a training set of 3464 MR brain images. Using these multi-atlas estimates we (1) estimate a low-dimensional representation for selecting locally appropriate example images, and (2) build AdaBoost learners that map a weak initial segmentation to the multi-atlas segmentation result. Thus, to segment a new target image we project the image into the low-dimensional space, construct a weak initial segmentation, and fuse the trained, locally selected, learners. The MLF framework cuts the runtime on a modern computer from 36 h down to 3-8 min - a 270× speedup - by completely bypassing the need for deformable atlas-target registrations. Additionally, we (1) describe a technique for optimizing the weak initial segmentation and the AdaBoost learning parameters, (2) quantify the ability to replicate the multi-atlas result with mean accuracies approaching the multi-atlas intra-subject reproducibility on a testing set of 380 images, (3) demonstrate significant increases in the reproducibility of intra-subject segmentations when compared to a state-of-the-art multi-atlas framework on a separate reproducibility dataset, (4) show that under the MLF framework the large-scale data model significantly improve the segmentation over the small-scale model under the MLF framework, and (5) indicate that the MLF framework has comparable performance as state-of-the-art multi-atlas segmentation algorithms without using non-local information.

Automatic tissue segmentation of head and neck MR images for hyperthermia treatment planning

A hyperthermia treatment requires accurate, patient-specific treatment planning. This planning is based on 3D anatomical models which are generally derived from computed tomography. Because of its superior soft tissue contrast, magnetic resonance imaging (MRI) information can be introduced to improve the quality of these 3D patient models and therefore the treatment planning itself. Thus, we present here an automatic atlas-based segmentation algorithm for MR images of the head and neck. Our method combines multiatlas local weighting fusion with intensity modelling. The accuracy of the method was evaluated using a leave-one-out cross validation experiment over a set of 11 patients for which manual delineation were available. The accuracy of the proposed method was high both in terms of the Dice similarity coefficient (DSC) and the 95th percentile Hausdorff surface distance (HSD) with median DSC higher than 0.8 for all tissues except sclera. For all tissues, except the spine tissues, the accuracy was approaching the interobserver agreement/variability both in terms of DSC and HSD. The positive effect of adding the intensity modelling to the multiatlas fusion decreased when a more accurate atlas fusion method was used.Using the proposed approach we improved the performance of the approach previously presented for H&N hyperthermia treatment planning, making the method suitable for clinical application.

Metastatic liver tumour segmentation from discriminant Grassmannian manifolds

The early detection, diagnosis and monitoring of liver cancer progression can be achieved with the precise delineation of metastatic tumours. However, accurate automated segmentation remains challenging due to the presence of noise, inhomogeneity and the high appearance variability of malignant tissue. In this paper, we propose an unsupervised metastatic liver tumour segmentation framework using a machine learning approach based on discriminant Grassmannian manifolds which learns the appearance of tumours with respect to normal tissue. First, the framework learns within-class and between-class similarity distributions from a training set of images to discover the optimal manifold discrimination between normal and pathological tissue in the liver. Second, a conditional optimisation scheme computes non-local pairwise as well as pattern-based clique potentials from the manifold subspace to recognise regions with similar labelings and to incorporate global consistency in the segmentation process. The proposed framework was validated on a clinical database of 43 CT images from patients with metastatic liver cancer. Compared to state-of-the-art methods, our method achieves a better performance on two separate datasets of metastatic liver tumours from different clinical sites, yielding an overall mean Dice similarity coefficient of [Formula: see text] in over 50 tumours with an average volume of 27.3 mm(3).

An Optimized PatchMatch for multi-scale and multi-feature label fusion

Automatic segmentation methods are important tools for quantitative analysis of Magnetic Resonance Images (MRI). Recently, patch-based label fusion approaches have demonstrated state-of-the-art segmentation accuracy. In this paper, we introduce a new patch-based label fusion framework to perform segmentation of anatomical structures. The proposed approach uses an Optimized PAtchMatch Label fusion (OPAL) strategy that drastically reduces the computation time required for the search of similar patches. The reduced computation time of OPAL opens the way for new strategies and facilitates processing on large databases. In this paper, we investigate new perspectives offered by OPAL, by introducing a new multi-scale and multi-feature framework. During our validation on hippocampus segmentation we use two datasets: young adults in the ICBM cohort and elderly adults in the EADC-ADNI dataset. For both, OPAL is compared to state-of-the-art methods. Results show that OPAL obtained the highest median Dice coefficient (89.9% for ICBM and 90.1% for EADC-ADNI). Moreover, in both cases, OPAL produced a segmentation accuracy similar to inter-expert variability. On the EADC-ADNI dataset, we compare the hippocampal volumes obtained by manual and automatic segmentation. The volumes appear to be highly correlated that enables to perform more accurate separation of pathological populations.

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.

Multi-Modality Vertebra Recognition in Arbitrary Views Using 3D Deformable Hierarchical Model

Computer-aided diagnosis of spine problems relies on the automatic identification of spine structures in images. The task of automatic vertebra recognition is to identify the global spine and local vertebra structural information such as spine shape, vertebra location and pose. Vertebra recognition is challenging due to the large appearance variations in different image modalities/views and the high geometric distortions in spine shape. Existing vertebra recognitions are usually simplified as vertebrae detections, which mainly focuses on the identification of vertebra locations and labels but cannot support further spine quantitative assessment. In this paper, we propose a vertebra recognition method using 3D deformable hierarchical model (DHM) to achieve cross-modality local vertebra location+pose identification with accurate vertebra labeling, and global 3D spine shape recovery. We recast vertebra recognition as deformable model matching, fitting the input spine images with the 3D DHM via deformations. The 3D model-matching mechanism provides a more comprehensive vertebra location+pose+label simultaneous identification than traditional vertebra location+label detection, and also provides an articulated 3D mesh model for the input spine section. Moreover, DHM can conduct versatile recognition on volume and multi-slice data, even on single slice. Experiments show our method can successfully extract vertebra locations, labels, and poses from multi-slice T1/T2 MR and volume CT, and can reconstruct 3D spine model on different image views such as lumbar, cervical, even whole spine. The resulting vertebra information and the recovered shape can be used for quantitative diagnosis of spine problems and can be easily digitalized and integrated in modern medical PACS systems.

Preliminary analysis using multi-atlas labeling algorithms for tracing longitudinal change

Multicenter longitudinal neuroimaging has great potential to provide efficient and consistent biomarkers for research of neurodegenerative diseases and aging. In rare disease studies it is of primary importance to have a reliable tool that performs consistently for data from many different collection sites to increase study power. A multi-atlas labeling algorithm is a powerful brain image segmentation approach that is becoming increasingly popular in image processing. The present study examined the performance of multi-atlas labeling tools for subcortical identification using two types of in-vivo image database: Traveling Human Phantom (THP) and PREDICT-HD. We compared the accuracy (Dice Similarity Coefficient; DSC and intraclass correlation; ICC), multicenter reliability (Coefficient of Variance; CV), and longitudinal reliability (volume trajectory smoothness and Akaike Information Criterion; AIC) of three automated segmentation approaches: two multi-atlas labeling tools, MABMIS and MALF, and a machine-learning-based tool, BRAINSCut. In general, MALF showed the best performance (higher DSC, ICC, lower CV, AIC, and smoother trajectory) with a couple of exceptions. First, the results of accumben, where BRAINSCut showed higher reliability, were still premature to discuss their reliability levels since their validity is still in doubt (DSC < 0.7, ICC < 0.7). For caudate, BRAINSCut presented slightly better accuracy while MALF showed significantly smoother longitudinal trajectory. We discuss advantages and limitations of these performance variations and conclude that improved segmentation quality can be achieved using multi-atlas labeling methods. While multi-atlas labeling methods are likely to help improve overall segmentation quality, caution has to be taken when one chooses an approach, as our results suggest that segmentation outcome can vary depending on research interest.

Automated hippocampal segmentation in 3D MRI using random undersampling with boosting algorithm

The automated identification of brain structure in Magnetic Resonance Imaging is very important both in neuroscience research and as a possible clinical diagnostic tool. In this study, a novel strategy for fully automated hippocampal segmentation in MRI is presented. It is based on a supervised algorithm, called RUSBoost, which combines data random undersampling with a boosting algorithm. RUSBoost is an algorithm specifically designed for imbalanced classification, suitable for large data sets because it uses random undersampling of the majority class. The RUSBoost performances were compared with those of ADABoost, Random Forest and the publicly available brain segmentation package, FreeSurfer. This study was conducted on a data set of 50 T1-weighted structural brain images. The RUSBoost-based segmentation tool achieved the best results with a Dice's index of [Formula: see text] ([Formula: see text]) for the left (right) brain hemisphere. An independent data set of 50 T1-weighted structural brain scans was used for an independent validation of the fully trained strategies. Again the RUSBoost segmentations compared favorably with manual segmentations with the highest performances among the four tools. Moreover, the Pearson correlation coefficient between hippocampal volumes computed by manual and RUSBoost segmentations was 0.83 (0.82) for left (right) side, statistically significant, and higher than those computed by Adaboost, Random Forest and FreeSurfer. The proposed method may be suitable for accurate, robust and statistically significant segmentations of hippocampi.