An automated registration algorithm for measuring MRI subcortical brain structures

Two-dimensional segmentation fusion tool: an extensible, free-to-use, user-friendly tool for combining different bidimensional segmentations

Introduction: In several fields, the process of fusing multiple two-dimensional (2D) closed lines is an important step. For instance, this is fundamental in histology and oncology in general. The treatment of a tumor consists of numerous steps and activities. Among them, segmenting the cancer area, that is, the correct identification of its spatial location by the segmentation technique, is one of the most important and at the same time complex and delicate steps. The difficulty in deriving reliable segmentations stems from the lack of a standard for identifying the edges and surrounding tissues of the tumor area. For this reason, the entire process is affected by considerable subjectivity. Given a tumor image, different practitioners can associate different segmentations with it, and the diagnoses produced may differ. Moreover, experimental data show that the analysis of the same area by the same physician at two separate timepoints may result in different lines being produced. Accordingly, it is challenging to establish which contour line is the ground truth. Methods: Starting from multiple segmentations related to the same tumor, statistical metrics and computational procedures could be exploited to combine them for determining the most reliable contour line. In particular, numerous algorithms have been developed over time for this procedure, but none of them is validated yet. Accordingly, in this field, there is no ground truth, and research is still active. Results: In this work, we developed the Two-Dimensional Segmentation Fusion Tool (TDSFT), a user-friendly tool distributed as a free-to-use standalone application for MAC, Linux, and Windows, which offers a simple and extensible interface where numerous algorithms are proposed to "compute the mean" (i.e., the process to fuse, combine, and "average") multiple 2D lines. Conclusions: The TDSFT can support medical specialists, but it can also be used in other fields where it is required to combine 2D close lines. In addition, the TDSFT is designed to be easily extended with new algorithms thanks to a dedicated graphical interface for configuring new parameters. The TDSFT can be downloaded from the following link: https://sourceforge.net/p/tdsft.

Comparative validation of AI and non-AI methods in MRI volumetry to diagnose Parkinsonian syndromes

Automated segmentation and volumetry of brain magnetic resonance imaging (MRI) scans are essential for the diagnosis of Parkinson's disease (PD) and Parkinson's plus syndromes (P-plus). To enhance the diagnostic performance, we adopt deep learning (DL) models in brain MRI segmentation and compared their performance with the gold-standard non-DL method. We collected brain MRI scans of healthy controls ([Formula: see text]) and patients with PD ([Formula: see text]), multiple systemic atrophy ([Formula: see text]), and progressive supranuclear palsy ([Formula: see text]) at Samsung Medical Center from January 2017 to December 2020. Using the gold-standard non-DL model, FreeSurfer (FS), we segmented six brain structures: midbrain, pons, caudate, putamen, pallidum, and third ventricle, and considered them as annotated data for DL models, the representative convolutional neural network (CNN) and vision transformer (ViT)-based models. Dice scores and the area under the curve (AUC) for differentiating normal, PD, and P-plus cases were calculated to determine the measure to which FS performance can be reproduced as-is while increasing speed by the DL approaches. The segmentation times of CNN and ViT for the six brain structures per patient were 51.26 ± 2.50 and 1101.82 ± 22.31 s, respectively, being 14 to 300 times faster than FS (15,735 ± 1.07 s). Dice scores of both DL models were sufficiently high (> 0.85) so their AUCs for disease classification were not inferior to that of FS. For classification of normal vs. P-plus and PD vs. P-plus (except multiple systemic atrophy - Parkinsonian type) based on all brain parts, the DL models and FS showed AUCs above 0.8, demonstrating the clinical value of DL models in addition to FS. DL significantly reduces the analysis time without compromising the performance of brain segmentation and differential diagnosis. Our findings may contribute to the adoption of DL brain MRI segmentation in clinical settings and advance brain research.

Integrated 3d flow-based multi-atlas brain structure segmentation

MRI brain structure segmentation plays an important role in neuroimaging studies. Existing methods either spend much CPU time, require considerable annotated data, or fail in segmenting volumes with large deformation. In this paper, we develop a novel multi-atlas-based algorithm for 3D MRI brain structure segmentation. It consists of three modules: registration, atlas selection and label fusion. Both registration and label fusion leverage an integrated flow based on grayscale and SIFT features. We introduce an effective and efficient strategy for atlas selection by employing the accompanying energy generated in the registration step. A 3D sequential belief propagation method and a 3D coarse-to-fine flow matching approach are developed in both registration and label fusion modules. The proposed method is evaluated on five public datasets. The results show that it has the best performance in almost all the settings compared to competitive methods such as ANTs, Elastix, Learning to Rank and Joint Label Fusion. Moreover, our registration method is more than 7 times as efficient as that of ANTs SyN, while our label transfer method is 18 times faster than Joint Label Fusion in CPU time. The results on the ADNI dataset demonstrate that our method is applicable to image pairs that require a significant transformation in registration. The performance on a composite dataset suggests that our method succeeds in a cross-modality manner. The results of this study show that the integrated 3D flow-based method is effective and efficient for brain structure segmentation. It also demonstrates the power of SIFT features, multi-atlas segmentation and classical machine learning algorithms for a medical image analysis task. The experimental results on public datasets show the proposed method’s potential for general applicability in various brain structures and settings.

Automated atlas-based segmentation for skull base surgical planning

PURPOSE

Computational surgical planning tools could help develop novel skull base surgical approaches that improve safety and patient outcomes. This defines a need for automated skull base segmentation to improve the usability of surgical planning software. The objective of this work was to design and validate an algorithm for atlas-based automated segmentation of skull base structures in individual image sets for skull base surgical planning.

METHODS

Advanced Normalization Tools software was used to construct a synthetic CT template from 6 subjects, and skull base structures were manually segmented to create a reference atlas. Landmark registration followed by Elastix deformable registration was applied to the template to register it to each of the 30 trusted reference image sets. Dice coefficient, average Hausdorff distance, and clinical usability scoring were used to compare the atlas segmentations to those of the trusted reference image sets.

RESULTS

The mean for average Hausdorff distance for all structures was less than 2 mm (mean for 95th percentile Hausdorff distance was less than 5 mm). For structures greater than 2.5 mL in volume, the average Dice coefficient was 0.73 (range 0.59-0.82), and for structures less than 2.5 mL in volume the Dice coefficient was less than 0.7. The usability scoring survey was completed by three experts, and all structures met the criteria for acceptable effort except for the foramen spinosum, rotundum, and carotid artery, which required more than minor corrections.

CONCLUSION

Currently available open-source algorithms, such as the Elastix deformable algorithm, can be used for automated atlas-based segmentation of skull base structures with acceptable clinical accuracy and minimal corrections with the use of the proposed atlas. The first publicly available CT template and anterior skull base segmentation atlas being released (available at this link: http://hdl.handle.net/1773/46259 ) with this paper will allow for general use of automated atlas-based segmentation of the skull base.

Automated segmentation of the hypothalamus and associated subunits in brain MRI

•

A publicly available deep learning tool to segment the hypothalamus and its subunits.

•

Our tool outperforms inter-rater accuracy and approaches intra-rater precision level.

•

It can robustly generalise to unseen heterogeneous datasets.

•

It yields a rejection rate of less than 1% in a QC analysis performed on 675 scans.

•

It detects subtle subunit-specific hypothalamic atrophy in Alzheimer’s Disease.

Despite the crucial role of the hypothalamus in the regulation of the human body, neuroimaging studies of this structure and its nuclei are scarce. Such scarcity partially stems from the lack of automated segmentation tools, since manual delineation suffers from scalability and reproducibility issues. Due to the small size of the hypothalamus and the lack of image contrast in its vicinity, automated segmentation is difficult and has been long neglected by widespread neuroimaging packages like FreeSurfer or FSL. Nonetheless, recent advances in deep machine learning are enabling us to tackle difficult segmentation problems with high accuracy. In this paper we present a fully automated tool based on a deep convolutional neural network, for the segmentation of the whole hypothalamus and its subregions from T1-weighted MRI scans. We use aggressive data augmentation in order to make the model robust to T1-weighted MR scans from a wide array of different sources, without any need for preprocessing. We rigorously assess the performance of the presented tool through extensive analyses, including: inter- and intra-rater variability experiments between human observers; comparison of our tool with manual segmentation; comparison with an automated method based on multi-atlas segmentation; assessment of robustness by quality control analysis of a larger, heterogeneous dataset (ADNI); and indirect evaluation with a volumetric study performed on ADNI. The presented model outperforms multi-atlas segmentation scores as well as inter-rater accuracy level, and approaches intra-rater precision. Our method does not require any preprocessing and runs in less than a second on a GPU, and approximately 10 seconds on a CPU. The source code as well as the trained model are publicly available at https://github.com/BBillot/hypothalamus_seg, and will also be distributed with FreeSurfer.

A Fast Subpixel Registration Algorithm Based on Single-Step DFT Combined with Phase Correlation Constraint in Multimodality Brain Image

Multimodality brain image registration technology is the key technology to determine the accuracy and speed of brain diagnosis and treatment. In order to achieve high-precision image registration, a fast subpixel registration algorithm based on single-step DFT combined with phase correlation constraint in multimodality brain image was proposed in this paper. Firstly, the coarse positioning at the pixel level was achieved by using the downsampling cross-correlation model, which reduced the Fourier transform dimension of the cross-correlation matrix and the multiplication of the discrete Fourier transform matrix, so as to speed up the coarse registration process. Then, the improved DFT multiplier of the matrix multiplication was used in the neighborhood of the coarse point, and the subpixel fast location was achieved by the bidirectional search strategy. Qualitative and quantitative simulation experiment results show that, compared with comparison registration algorithms, our proposed algorithm could greatly reduce space and time complexity without losing accuracy.

Quantifying deep grey matter atrophy using automated segmentation approaches: A systematic review of structural MRI studies

The deep grey matter (DGM) nuclei of the brain play a crucial role in learning, behaviour, cognition, movement and memory. Although automated segmentation strategies can provide insight into the impact of multiple neurological conditions affecting these structures, such as Multiple Sclerosis (MS), Huntington's disease (HD), Alzheimer's disease (AD), Parkinson's disease (PD) and Cerebral Palsy (CP), there are a number of technical challenges limiting an accurate automated segmentation of the DGM. Namely, the insufficient contrast of T1 sequences to completely identify the boundaries of these structures, as well as the presence of iso-intense white matter lesions or extensive tissue loss caused by brain injury. Therefore in this systematic review, 269 eligible studies were analysed and compared to determine the optimal approaches for addressing these technical challenges. The automated approaches used among the reviewed studies fall into three broad categories, atlas-based approaches focusing on the accurate alignment of atlas priors, algorithmic approaches which utilise intensity information to a greater extent, and learning-based approaches that require an annotated training set. Studies that utilise freely available software packages such as FIRST, FreeSurfer and LesionTOADS were also eligible, and their performance compared. Overall, deep learning approaches achieved the best overall performance, however these strategies are currently hampered by the lack of large-scale annotated data. Improving model generalisability to new datasets could be achieved in future studies with data augmentation and transfer learning. Multi-atlas approaches provided the second-best performance overall, and may be utilised to construct a "silver standard" annotated training set for deep learning. To address the technical challenges, providing robustness to injury can be improved by using multiple channels, highly elastic diffeomorphic transformations such as LDDMM, and by following atlas-based approaches with an intensity driven refinement of the segmentation, which has been done with the Expectation Maximisation (EM) and level sets methods. Accounting for potential lesions should be achieved with a separate lesion segmentation approach, as in LesionTOADS. Finally, to address the issue of limited contrast, R2*, T2* and QSM sequences could be used to better highlight the DGM due to its higher iron content. Future studies could look to additionally acquire these sequences by retaining the phase information from standard structural scans, or alternatively acquiring these sequences for only a training set, allowing models to learn the "improved" segmentation from T1-sequences alone.

A fast atlas pre-selection procedure for multi-atlas based brain segmentation

Multi-atlas based MR image segmentation has been recognized as a quantitative analysis approach for brain. For such purpose, atlas databases keep increasing to include various anatomical characteristics of human brain. Atlas pre-selection becomes a necessary step for efficient and accurate automated segmentation of human brain images. In this study, we proposed a method of atlas pre-selection for target image segmentation on the MriCloud platform, which is a state-of-the-art multi-atlas based segmentation tool. In the MRIcloud pipeline, segmentation of lateral ventricle (LV) label is generated as an additional input in the segmentation pipeline. Under this circumstance, similarity of the LV label between target image and atlases was adopted as the atlas ranking scheme. Dice overlap coefficient was calculated and taken as the quantitative measure for atlas ranking. Segmentation results based on the proposed method were compared with that based on atlas pre-selection by mutual information (MI) between images. The final segmentation results showed a comparable accuracy of the proposed method with that from MI based atlas pre-selection. However, the computation load for the atlas pre-selection was speeded up by about 20 times compared to MI based pre-selection. The proposed method provides a promising assistance for quantitative analysis of brain images.

Automatic segmentation of the striatum and globus pallidus using MIST: Multimodal Image Segmentation Tool

Accurate segmentation of the subcortical structures is frequently required in neuroimaging studies. Most existing methods use only a T1-weighted MRI volume to segment all supported structures and usually rely on a database of training data. We propose a new method that can use multiple image modalities simultaneously and a single reference segmentation for initialisation, without the need for a manually labelled training set. The method models intensity profiles in multiple images around the boundaries of the structure after nonlinear registration. It is trained using a set of unlabelled training data, which may be the same images that are to be segmented, and it can automatically infer the location of the physical boundary using user-specified priors. We show that the method produces high-quality segmentations of the striatum, which is clearly visible on T1-weighted scans, and the globus pallidus, which has poor contrast on such scans. The method compares favourably to existing methods, showing greater overlap with manual segmentations and better consistency.

Use of T1-weighted/T2-weighted magnetic resonance ratio images to elucidate changes in the schizophrenic brain

INTRODUCTION

One leading hypothesis suggests that schizophrenia (SZ) is a neurodevelopmental disorder caused by genetic defects in association with environmental risk factors that affect synapse and myelin formation. Recent magnetic resonance imaging (MRI) studies of SZ brain showed both gray matter (GM) reduction and white matter (WM) fractional anisotropy reduction. In this study, we used T1-weighted (T1w)/T2-weighted (T2w) MRI ratio images, which increase myelin-related signal contrast and reduce receiver-coil bias.

METHODS

We measured T1w/T2w ratio image signal intensity in 29 patients with SZ and 33 healthy controls (HCs), and then compared them against bias-corrected T1w images.

RESULTS

Mean T1w/T2w ratio signal intensity values across all SZ GM and WM voxels were significantly lower than those for the HC values (analysis of covariance with age, gender, handedness, and premorbid intelligence quotient as nuisance covariates). SZ mean WM T1w/T2w ratio values were related to Global Assessment of Functioning (GAF) scores and were inversely related to the positive psychotic symptoms of the Positive and Negative Syndrome Scale. Voxel-based analysis revealed significantly lower T1w/T2w ratio image signal intensity values in the right ventral putamen in SZ GM. T1w image intensities did not differ between the SZ and HC groups.

CONCLUSIONS

T1-weighted/T2-weighted ratio imaging increased the detectability of SZ pathological changes. Reduced SZ brain signal intensity is likely due to diminished myelin content; therefore, mapping myelin-related SZ brain changes using T1w/T2w ratio images may be useful for studies of SZ brain abnormalities.

Content-based image retrieval for brain MRI: an image-searching engine and population-based analysis to utilize past clinical data for future diagnosis

Radiological diagnosis is based on subjective judgment by radiologists. The reasoning behind this process is difficult to document and share, which is a major obstacle in adopting evidence-based medicine in radiology. We report our attempt to use a comprehensive brain parcellation tool to systematically capture image features and use them to record, search, and evaluate anatomical phenotypes. Anatomical images (T1-weighted MRI) were converted to a standardized index by using a high-dimensional image transformation method followed by atlas-based parcellation of the entire brain. We investigated how the indexed anatomical data captured the anatomical features of healthy controls and a population with Primary Progressive Aphasia (PPA). PPA was chosen because patients have apparent atrophy at different degrees and locations, thus the automated quantitative results can be compared with trained clinicians' qualitative evaluations. We explored and tested the power of individual classifications and of performing a search for images with similar anatomical features in a database using partial least squares-discriminant analysis (PLS-DA) and principal component analysis (PCA). The agreement between the automated z-score and the averaged visual scores for atrophy (r = 0.8) was virtually the same as the inter-evaluator agreement. The PCA plot distribution correlated with the anatomical phenotypes and the PLS-DA resulted in a model with an accuracy of 88% for distinguishing PPA variants. The quantitative indices captured the main anatomical features. The indexing of image data has a potential to be an effective, comprehensive, and easily translatable tool for clinical practice, providing new opportunities to mine clinical databases for medical decision support.

•

Brain parcellation tools define structures automatically and convert images into standardized and quantitative matrices.

•

We tested if an automated tool and the resultant vector of structural volumes can accurately capture anatomical phenotypes.

•

The agreement between visual and automated atrophy detection was virtually the same as the inter-evaluator agreement.

•

The quantitative indices captured the main anatomical features in brains with atrophy in different degrees and location.

•

The image quantification has potential to be an effective, comprehensive, and easily translatable tool for clinical practice.

Automated algorithm to measure changes in medial temporal lobe volume in Alzheimer disease

BACKGROUND

The change in volume of anatomic structures is as a sensitive indicator of Alzheimer disease (AD) progression. Although several methods are available to measure brain volumes, improvements in speed and automation are required. Our objective was to develop a fully automated, fast, and reliable approach to measure change in medial temporal lobe (MTL) volume, including primarily hippocampus.

METHODS

The MTL volume defined in an atlas image was propagated onto each baseline image and a level set algorithm was applied to refine the shape and smooth the boundary. The MTL of the baseline image was then mapped onto the corresponding follow-up image to measure volume change (ΔMTL). Baseline and 24 months 3D T1-weighted images from the Alzheimer Disease Neuroimaging Initiative (ADNI) were randomly selected for 50 normal elderly controls (NECs), 50 subjects with mild cognitive impairment (MCI) and 50 subjects with AD to test the algorithm. The method was compared to the FreeSurfer segmentation tools.

RESULTS

The average ΔMTL (mean±SEM) was 68±35mm(3) in NEC, 187±38mm(3) in MCI and 300±34mm(3) in the AD group and was significantly different (p<0.0001) between all three groups. The ΔMTL was correlated with cognitive decline.

COMPARISON WITH EXISTING METHOD(S)

Results for the FreeSurfer software were similar but did not detect significant differences between the MCI and AD groups.

CONCLUSION

This novel segmentation approach is fully automated and provides a robust marker of brain atrophy that shows different rates of atrophy over 2 years between NEC, MCI, and AD groups.

Automatic hippocampus segmentation of 7.0 Tesla MR images by combining multiple atlases and auto-context models

In many neuroscience and clinical studies, accurate measurement of hippocampus is very important to reveal the inter-subject anatomical differences or the subtle intra-subject longitudinal changes due to aging or dementia. Although many automatic segmentation methods have been developed, their performances are still challenged by the poor image contrast of hippocampus in the MR images acquired especially from 1.5 or 3.0 Tesla (T) scanners. With the recent advance of imaging technology, 7.0 T scanner provides much higher image contrast and resolution for hippocampus study. However, the previous methods developed for segmentation of hippocampus from 1.5 T or 3.0 T images do not work for the 7.0 T images, due to different levels of imaging contrast and texture information. In this paper, we present a learning-based algorithm for automatic segmentation of hippocampi from 7.0 T images, by taking advantages of the state-of-the-art multi-atlas framework and also the auto-context model (ACM). Specifically, ACM is performed in each atlas domain to iteratively construct sequences of location-adaptive classifiers by integrating both image appearance and local context features. Due to the plenty texture information in 7.0 T images, more advanced texture features are also extracted and incorporated into the ACM during the training stage. Then, under the multi-atlas segmentation framework, multiple sequences of ACM-based classifiers are trained for all atlases to incorporate the anatomical variability. In the application stage, for a new image, its hippocampus segmentation can be achieved by fusing the labeling results from all atlases, each of which is obtained by applying the atlas-specific ACM-based classifiers. Experimental results on twenty 7.0 T images with the voxel size of 0.35×0.35×0.35 mm3 show very promising hippocampus segmentations (in terms of Dice overlap ratio 89.1±0.020), indicating high applicability for the future clinical and neuroscience studies.

Local label learning (LLL) for subcortical structure segmentation: application to hippocampus segmentation

Automatic and reliable segmentation of subcortical structures is an important but difficult task in quantitative brain image analysis. Multi-atlas based segmentation methods have attracted great interest due to their promising performance. Under the multi-atlas based segmentation framework, using deformation fields generated for registering atlas images onto a target image to be segmented, labels of the atlases are first propagated to the target image space and then fused to get the target image segmentation based on a label fusion strategy. While many label fusion strategies have been developed, most of these methods adopt predefined weighting models that are not necessarily optimal. In this study, we propose a novel local label learning strategy to estimate the target image's segmentation label using statistical machine learning techniques. In particular, we use a L1-regularized support vector machine (SVM) with a k nearest neighbor (kNN) based training sample selection strategy to learn a classifier for each of the target image voxel from its neighboring voxels in the atlases based on both image intensity and texture features. Our method has produced segmentation results consistently better than state-of-the-art label fusion methods in validation experiments on hippocampal segmentation of over 100 MR images obtained from publicly available and in-house datasets. Volumetric analysis has also demonstrated the capability of our method in detecting hippocampal volume changes due to Alzheimer's disease.

Mapping registration sensitivity in MR mouse brain images

Nonlinear registration algorithms provide a way to estimate structural (brain) differences based on magnetic resonance images. Their ability to align images of different individuals and across modalities has been well-researched, but the bounds of their sensitivity with respect to the recovery of salient morphological differences between groups are unclear. Here we develop a novel approach to simulate deformations on MR brain images to evaluate the ability of two registration algorithms to extract structural differences corresponding to biologically plausible atrophy and expansion. We show that at a neuroanatomical level registration accuracy is influenced by the size and compactness of structures, but do so differently depending on how much change is simulated. The size of structures has a small influence on the recovered accuracy. There is a trend for larger structures to be recovered more accurately, which becomes only significant as the amount of simulated change is large. More compact structures can be recovered more accurately regardless of the amount of simulated change. Both tested algorithms underestimate the full extent of the simulated atrophy and expansion. Finally we show that when multiple comparisons are corrected for at a voxelwise level, a very low rate of false positives is obtained. More interesting is that true positive rates average around 40%, indicating that the simulated changes are not fully recovered. Simulation experiments were run using two fundamentally different registration algorithms and we identified the same results, suggesting that our findings are generalizable across different classes of nonlinear registration algorithms.

Magnetic resonance imaging of the newborn brain: automatic segmentation of brain images into 50 anatomical regions

We studied methods for the automatic segmentation of neonatal and developing brain images into 50 anatomical regions, utilizing a new set of manually segmented magnetic resonance (MR) images from 5 term-born and 15 preterm infants imaged at term corrected age called ALBERTs. Two methods were compared: individual registrations with label propagation and fusion; and template based registration with propagation of a maximum probability neonatal ALBERT (MPNA). In both cases we evaluated the performance of different neonatal atlases and MPNA, and the approaches were compared with the manual segmentations by means of the Dice overlap coefficient. Dice values, averaged across regions, were 0.81±0.02 using label propagation and fusion for the preterm population, and 0.81±0.02 using the single registration of a MPNA for the term population. Segmentations of 36 further unsegmented target images of developing brains yielded visibly high-quality results. This registration approach allows the rapid construction of automatically labeled age-specific brain atlases for neonates and the developing brain.

Diffeomorphic brain mapping based on T1-weighted images: improvement of registration accuracy by multichannel mapping

PURPOSE

To improve image registration accuracy in neurodegenerative populations.

MATERIALS AND METHODS

This study used primary progressive aphasia, aged control, and young control T1-weighted images. Mapping to a template image was performed using single-channel Large Deformation Diffeomorphic Metric Mapping (LDDMM), a dual-channel method with ventricular anatomy in the second channel, and a dual-channel with appendage method, which utilized a priori knowledge of template ventricular anatomy in the deformable atlas.

RESULTS

Our results indicated substantial improvement in the registration accuracy over single-contrast-based brain mapping, mainly in the lateral ventricles and regions surrounding them. Dual-channel mapping significantly (P < 0.001) reduced the number of misclassified lateral ventricle voxels (based on a manually defined reference) over single-channel mapping. The dual-channel (w/appendage) method further reduced (P < 0.001) misclassification over the dual-channel method, indicating that the appendage provides more accurate anatomical correspondence for deformation.

CONCLUSION

Brain anatomical mapping by shape normalization is widely used for quantitative anatomical analysis. However, in many geriatric and neurodegenerative disorders, severe tissue atrophy poses a unique challenge for accurate mapping of voxels, especially around the lateral ventricles. In this study we demonstrate our ability to improve mapping accuracy by incorporating ventricular anatomy in LDDMM and by utilizing a priori knowledge of ventricular anatomy in the deformable atlas.

Atlas-based automatic mouse brain image segmentation revisited: model complexity vs. image registration

Although many atlas-based segmentation methods have been developed and validated for the human brain, limited work has been done for the mouse brain. This paper investigated roles of image registration and segmentation model complexity in the mouse brain segmentation. We employed four segmentation models [single atlas, multiatlas, simultaneous truth and performance level estimation (STAPLE) and Markov random field (MRF) via four different image registration algorithms (affine, B-spline free-form deformation (FFD), Demons and large deformation diffeomorphic metric mapping (LDDMM)] for delineating 19 structures from in vivo magnetic resonance microscopy images. We validated their accuracies against manual segmentation. Our results revealed that LDDMM outperformed Demons, FFD and affine in any of the segmentation models. Under the same registration, increasing segmentation model complexity from single atlas to multiatlas, STAPLE or MRF significantly improved the segmentation accuracy. Interestingly, the multiatlas-based segmentation using nonlinear registrations (FFD, Demons and LDDMM) had similar performance to their STAPLE counterparts, while they both outperformed their MRF counterparts. Furthermore, when the single-atlas affine segmentation was used as reference, the improvement due to nonlinear registrations (FFD, Demons and LDDMM) in the single-atlas segmentation model was greater than that due to increasing model complexity (multiatlas, STAPLE and MRF affine segmentation). Hence, we concluded that image registration plays a more crucial role in the atlas-based automatic mouse brain segmentation as compared to model complexity. Multiple atlases with LDDMM can best improve the segmentation accuracy in the mouse brain among all segmentation models tested in this study.

A fully-automatic caudate nucleus segmentation of brain MRI: application in volumetric analysis of pediatric attention-deficit/hyperactivity disorder

Background

Accurate automatic segmentation of the caudate nucleus in magnetic resonance images (MRI) of the brain is of great interest in the analysis of developmental disorders. Segmentation methods based on a single atlas or on multiple atlases have been shown to suitably localize caudate structure. However, the atlas prior information may not represent the structure of interest correctly. It may therefore be useful to introduce a more flexible technique for accurate segmentations.

Method

We present Cau-dateCut: a new fully-automatic method of segmenting the caudate nucleus in MRI. CaudateCut combines an atlas-based segmentation strategy with the Graph Cut energy-minimization framework. We adapt the Graph Cut model to make it suitable for segmenting small, low-contrast structures, such as the caudate nucleus, by defining new energy function data and boundary potentials. In particular, we exploit information concerning the intensity and geometry, and we add supervised energies based on contextual brain structures. Furthermore, we reinforce boundary detection using a new multi-scale edgeness measure.

Results

We apply the novel CaudateCut method to the segmentation of the caudate nucleus to a new set of 39 pediatric attention-deficit/hyperactivity disorder (ADHD) patients and 40 control children, as well as to a public database of 18 subjects. We evaluate the quality of the segmentation using several volumetric and voxel by voxel measures. Our results show improved performance in terms of segmentation compared to state-of-the-art approaches, obtaining a mean overlap of 80.75%. Moreover, we present a quantitative volumetric analysis of caudate abnormalities in pediatric ADHD, the results of which show strong correlation with expert manual analysis.

Conclusion

CaudateCut generates segmentation results that are comparable to gold-standard segmentations and which are reliable in the analysis of differentiating neuroanatomical abnormalities between healthy controls and pediatric ADHD.

A review of atlas-based segmentation for magnetic resonance brain images

Normal and abnormal brains can be segmented by registering the target image with an atlas. Here, an atlas is defined as the combination of an intensity image (template) and its segmented image (the atlas labels). After registering the atlas template and the target image, the atlas labels are propagated to the target image. We define this process as atlas-based segmentation. In recent years, researchers have investigated registration algorithms to match atlases to query subjects and also strategies for atlas construction. In this paper we present a review of the automated approaches for atlas-based segmentation of magnetic resonance brain images. We aim to point out the strengths and weaknesses of atlas-based methods and suggest new research directions. We use two different criteria to present the methods. First, we refer to the algorithms according to their atlas-based strategy: label propagation, multi-atlas methods, and probabilistic techniques. Subsequently, we classify the methods according to their medical target: the brain and its internal structures, tissue segmentation in healthy subjects, tissue segmentation in fetus, neonates and elderly subjects, and segmentation of damaged brains. A quantitative comparison of the results reported in the literature is also presented.

A generative model for image segmentation based on label fusion

We propose a nonparametric, probabilistic model for the automatic segmentation of medical images, given a training set of images and corresponding label maps. The resulting inference algorithms rely on pairwise registrations between the test image and individual training images. The training labels are then transferred to the test image and fused to compute the final segmentation of the test subject. Such label fusion methods have been shown to yield accurate segmentation, since the use of multiple registrations captures greater inter-subject anatomical variability and improves robustness against occasional registration failures. To the best of our knowledge, this manuscript presents the first comprehensive probabilistic framework that rigorously motivates label fusion as a segmentation approach. The proposed framework allows us to compare different label fusion algorithms theoretically and practically. In particular, recent label fusion or multiatlas segmentation algorithms are interpreted as special cases of our framework. We conduct two sets of experiments to validate the proposed methods. In the first set of experiments, we use 39 brain MRI scans-with manually segmented white matter, cerebral cortex, ventricles and subcortical structures-to compare different label fusion algorithms and the widely-used FreeSurfer whole-brain segmentation tool. Our results indicate that the proposed framework yields more accurate segmentation than FreeSurfer and previous label fusion algorithms. In a second experiment, we use brain MRI scans of 282 subjects to demonstrate that the proposed segmentation tool is sufficiently sensitive to robustly detect hippocampal volume changes in a study of aging and Alzheimer's Disease.

The impact of brain size on pilot performance varies with aviation training and years of education

Previous studies have consistently reported age-related changes in cognitive abilities and brain structure. Previous studies also suggest compensatory roles for specialized training, skill, and years of education in the age-related decline of cognitive function. The Stanford/VA Aviation Study examines the influence of specialized training and skill level (expertise) on age-related changes in cognition and brain structure. This preliminary report examines the effect of aviation expertise, years of education, age, and brain size on flight simulator performance in pilots aged 45-68 years. Fifty-one pilots were studied with structural magnetic resonance imaging, flight simulator, and processing speed tasks. There were significant main effects of age (p < .01) and expertise (p < .01), but not of whole brain size (p > .1) or education (p > .1), on flight simulator performance. However, even though age and brain size were correlated (r = -0.41), age differences in flight simulator performance were not explained by brain size. Both aviation expertise and education were involved in an interaction with brain size in predicting flight simulator performance (p < .05). These results point to the importance of examining measures of expertise and their interactions to assess age-related cognitive changes.

Reliability and validity of MRI-based automated volumetry software relative to auto-assisted manual measurement of subcortical structures in HIV-infected patients from a multisite study

The automated volumetric output of FreeSurfer and Individual Brain Atlases using Statistical Parametric Mapping (IBASPM), two widely used and well published software packages, was examined for accuracy and consistency relative to auto-assisted manual (AAM) tracings (i.e., manual correction of automated output) when measuring the caudate, putamen, amygdala, and hippocampus in the baseline scans of 120 HIV-infected patients (86.7% male, 47.3+/-6.3y.o., mean HIV duration 12.0+/-6.3years) from the NIH-funded HIV Neuroimaging Consortium (HIVNC) cohort. The data was examined for accuracy and consistency relative to auto-assisted manual tracing, and construct validity was assessed by correlating automated and AAM volumetric measures with relevant clinical measures of HIV progression. When results were averaged across all patients in the eight structures examined, FreeSurfer achieved lower absolute volume difference in five, higher sensitivity in seven, and higher spatial overlap in all eight structures. Additionally, FreeSurfer results exhibited less variability in all measures. Output from both methods identified discrepant correlations with clinical measures of HIV progression relative to AAM segmented data. Overall, FreeSurfer proved more effective in the context of subcortical volumetry in HIV-patients, particularly in a multisite cohort study such as this. These findings emphasize that regardless of the automated method used, visual inspection of segmentation output, along with manual correction if necessary, remains critical to ensuring the validity of reported results.

An evaluation of four automatic methods of segmenting the subcortical structures in the brain

The automation of segmentation of subcortical structures in the brain is an active research area. We have comprehensively evaluated four novel methods of fully automated segmentation of subcortical structures using volumetric, spatial overlap and distance-based measures. Two methods are atlas-based - classifier fusion and labelling (CFL) and expectation-maximisation segmentation using a brain atlas (EMS), and two incorporate statistical models of shape and appearance - profile active appearance models (PAM) and Bayesian appearance models (BAM). Each method was applied to the segmentation of 18 subcortical structures in 270 subjects from a diverse pool varying in age, disease, sex and image acquisition parameters. Our results showed that all four methods perform on par with recently published methods. CFL performed better than the others according to all three classes of metrics. In summary over all structures, the ranking by the Dice coefficient was CFL, BAM, joint EMS and PAM. The Hausdorff distance ranked the methods as CFL, joint PAM and BAM, EMS, whilst percentage absolute volumetric difference ranked them as joint CFL and PAM, joint BAM and EMS. Furthermore, as we had four methods of performing segmentation, we investigated whether the results obtained by each method were more similar to each other than to the manual segmentations using Williams' Index. Reassuringly, the Williams' Index was close to 1 for most subjects (mean=1.02, sd=0.05), indicating better agreement of each method with the gold standard than with the other methods. However, 2% of cases (mainly amygdala and nucleus accumbens) had values outside 3 standard deviations of the mean.

The emerging discipline of Computational Functional Anatomy

Computational Functional Anatomy (CFA) is the study of functional and physiological response variables in anatomical coordinates. For this we focus on two things: (i) the construction of bijections (via diffeomorphisms) between the coordinatized manifolds of human anatomy, and (ii) the transfer (group action and parallel transport) of functional information into anatomical atlases via these bijections. We review advances in the unification of the bijective comparison of anatomical submanifolds via point-sets including points, curves and surface triangulations as well as dense imagery. We examine the transfer via these bijections of functional response variables into anatomical coordinates via group action on scalars and matrices in DTI as well as parallel transport of metric information across multiple templates which preserves the inner product.

Multi-atlas based segmentation of brain images: atlas selection and its effect on accuracy

Quantitative research in neuroimaging often relies on anatomical segmentation of human brain MR images. Recent multi-atlas based approaches provide highly accurate structural segmentations of the brain by propagating manual delineations from multiple atlases in a database to a query subject and combining them. The atlas databases which can be used for these purposes are growing steadily. We present a framework to address the consequent problems of scale in multi-atlas segmentation. We show that selecting a custom subset of atlases for each query subject provides more accurate subcortical segmentations than those given by non-selective combination of random atlas subsets. Using a database of 275 atlases, we tested an image-based similarity criterion as well as a demographic criterion (age) in a leave-one-out cross-validation study. Using a custom ranking of the database for each subject, we combined a varying number n of atlases from the top of the ranked list. The resulting segmentations were compared with manual reference segmentations using Dice overlap. Image-based selection provided better segmentations than random subsets (mean Dice overlap 0.854 vs. 0.811 for the estimated optimal subset size, n=20). Age-based selection resulted in a similar marked improvement. We conclude that selecting atlases from large databases for atlas-based brain image segmentation improves the accuracy of the segmentations achieved. We show that image similarity is a suitable selection criterion and give results based on selecting atlases by age that demonstrate the value of meta-information for selection.

Non-rigid registration based segmentation of brain subcortical structures using a priori knowledge

Segmentation of the brain internal structures is an important and a challenging task due to their complex shapes, partial volume effects, low contrasts and anatomical variability between subjects. In this paper we propose a new non-rigid registration method that automatically segments the deep brain internal structures from brain MRI images. An atlas of the structures is used as a priori knowledge, which is modeled as a shape representation. By integrating the shape knowledge into a classical intensity based non-rigid registration algorithm, the proposed segmentation method allows to ameliorate the results in the case of low contrast on the boundaries of the structures. The shape model is based on distance representation obtained from the atlas. The segmentation of brain subcortical structures is performed on real MRI images and the obtained results are very encouraging.

Markov dependence tree-based segmentation of deep brain structures

We propose a new framework for multi-object segmentation of deep brain structures, which have significant shape variations and relatively small sizes in medical brain images. In the images, the structure boundaries may be blurry or even missing, and the surrounding background is a clutter and full of irrelevant edges. We suggest a template-based framework, which fuses the information of edge features, region statistics and inter-structure constraints to detect and locate all the targeted brain structures such that manual initialization is unnecessary. The multi-object template is organized in the form of a hierarchical Markov dependence tree. It makes the matching of multiple objects efficient. Our approach needs only one example as training data and alleviates the demand of a large training set. The obtained segmentation results on real data are encouraging and the proposed method enjoys several important advantages over existing methods.

An MRI study of age-related white and gray matter volume changes in the rhesus monkey

We applied the automated MRI segmentation technique Template Driven Segmentation (TDS) to dual-echo spin echo (DE SE) images of eight young (5-12 years), six middle-aged (16-19 years) and eight old (24-30 years) rhesus monkeys. We analyzed standardized mean volumes for 18 anatomically defined regions of interest (ROI's) and found an overall decrease from young to old age in the total forebrain (5.01%), forebrain parenchyma (5.24%), forebrain white matter (11.53%), forebrain gray matter (2.08%), caudate nucleus (11.79%) and globus pallidus (18.26%). Corresponding behavioral data for five of the young, five of the middle-aged and seven of the old subjects on the Delayed Non-matching to Sample (DNMS) task, the Delayed-recognition Span Task (DRST) and the Cognitive Impairment Index (CII) were also analyzed. We found that none of the cognitive measures were related to ROI volume changes in our sample size of monkeys.

Hippocampus segmentation in MR images using atlas registration, voxel classification, and graph cuts

Since hippocampal volume has been found to be an early biomarker for Alzheimer's disease, there is large interest in automated methods to accurately, robustly, and reproducibly extract the hippocampus from MRI data. In this work we present a segmentation method based on the minimization of an energy functional with intensity and prior terms, which are derived from manually labelled training images. The intensity energy is based on a statistical intensity model that is learned from the training images. The prior energy consists of a spatial and regularity term. The spatial prior is obtained from a probabilistic atlas created by registering the training images to the unlabelled target image, and deforming and averaging the training labels. The regularity prior energy encourages smooth segmentations. The resulting energy functional is globally minimized using graph cuts. The method was evaluated using image data from a population-based study on diseases among the elderly. Two set of images were used: a small set of 20 manually labelled MR images and a larger set of 498 images, for which manual volume measurements were available, but no segmentations. This data was previously used in a volumetry study that found significant associations between hippocampal volume and cognitive decline and incidence of dementia. Cross-validation experiments with the labelled set showed similarity indices of 0.852 and 0.864 and mean surface distances of 0.40 and 0.36 mm for the left and right hippocampus. 83% of the automated segmentations of the large set were rated as 'good' by a trained observer. Also, the proposed method was used to repeat the manual hippocampal volumetry study. The automatically obtained hippocampal volumes showed significant associations with cognitive decline and dementia, similar to the manually measured volumes. Finally, direct quantitative and qualitative comparisons showed that the proposed method outperforms a multi-atlas based segmentation method.

Effects of spatial transformation on regional brain volume estimates

Spatial transformation of MR brain images is a standard tool used in automated anatomical parcellation and other quantitative and qualitative methods to assess brain tissue volume, composition, and distribution. Despite widespread use, the quantitative effects of spatial transformation on regional brain volume estimates have been little studied. We report on the effects of transformation on regional brain volumes of 38 (17M, 21F) manually parcellated brains. After tracing in native space, regions of interest were transformed using a classic piecewise-linear Talairach transformation (Tal) or a nonlinear registration (AIR 5th order nonlinear algorithm, 158 parameters) to one of three Talairach-based templates: 1) Tal50, constructed from 50 Talairach-transformed normal brains, 2) the MNI 305 atlas, 3) IA38, constructed from MNI305-transformed scans of the 38 subjects used in this study. Native volumes were compared to the transformed volumes. We found that: 1) significant group-level differences can be obtained in transformed data sets that are in the opposite direction of effects obtained in native space; 2) the effects of transformation are heterogeneous across brain regions, even after covarying for total brain volume and age; 3) volumetric intra-class correlations between native and transformed brains differ by registration method and template choice, region, and tissue type; and 4) transformed brains produced hippocampus and corpus callosum volume proportions that were significantly different from those obtained in native space. Our results suggest that region-based volumetric differences uncovered by spatial-transformation-based methods should be replicated in native-space brains, and that meta-analyses should take into account whether volumes are determined using spatially-transformed images and/or specific automated methods.

A rhesus monkey reference label atlas for template driven segmentation

BACKGROUND

We have acquired dual-echo spin-echo (DE SE) MRI data of the rhesus monkey brain since 1994 as part of an ongoing study of normal aging. To analyze these legacy data for regional volume changes, we have created a reference label atlas for the Template Driven Segmentation (TDS) algorithm.

METHODS

The atlas was manually created from DE SE legacy MRI data of one behaviorally normal, young, male rhesus monkey and consisted of 14 regions of interest (ROI's). We analyzed the reproducibility and validity of the TDS algorithm using the atlas relative to manual segmentation.

RESULTS

ROI volumes were comparable between the two segmentation methodologies, except TDS overestimated the volume of basal ganglia regions. Both methodologies were highly reproducible, but TDS had lower sensitivity and comparable specificity.

CONCLUSIONS

TDS segmentation calculates accurate volumes for most ROI's. Sensitivity will be improved in future studies through the acquisition of higher quality data.

Registration and machine learning-based automated segmentation of subcortical and cerebellar brain structures

The large amount of imaging data collected in several ongoing multi-center studies requires automated methods to delineate brain structures of interest. We have previously reported on using artificial neural networks (ANN) to define subcortical brain structures. Here we present several automated segmentation methods using multidimensional registration. A direct comparison between template, probability, artificial neural network (ANN) and support vector machine (SVM)-based automated segmentation methods is presented. Three metrics for each segmentation method are reported in the delineation of subcortical and cerebellar brain regions. Results show that the machine learning methods outperform the template and probability-based methods. Utilization of these automated segmentation methods may be as reliable as manual raters and require no rater intervention.

Classifier selection strategies for label fusion using large atlas databases

Structural segmentations of brain MRI can be generated by propagating manually labelled atlas images from a repository to a query subject and combining them. This method has been shown to be robust, consistent and increasingly accurate with increasing numbers of classifiers. It outperforms standard atlas-based segmentation but suffers, however, from problems of scale when the number of atlases is large. For a large repository and a particular query subject, using a selection strategy to identify good classifiers is one way to address problems of scale. This work presents and compares different classifier selection strategies which are applied to a group of 275 subjects with manually labelled brain MR images. We approximate an upper limit for the accuracy or overlap that can be achieved for a particular structure in a given subject and compare this with the accuracy obtained using classifier selection. The accuracy of different classifier selection strategies are also rated against the distribution of overlaps generated by random groups of classifiers.

Development of an auxiliary system for the execution of vascular catheter interventions with a reduced radiological risk; system description and first experimental results

Vascular catheterization is a common procedure in clinical medicine. It is normally performed by a specialist using an X-ray fluoroscopic guide and contrast-media. In the present paper, an image-guided navigation system which indicates a path providing guidance to the desired target inside the vascular tree is described with the aim of reducing the exposure of personnel and patients to X-rays during the catheterization procedure. A 3D model of the patient vascular tree, reconstructed with data collected by an angiography before starting the intervention, is used as a guide map instead of fluoroscopic scans. An accurate spatial correspondence between the body of the patient and the 3D reconstructed vascular model is established and, by means of a position indicator installed over the catheter tip, the real-time position/orientation of the tip is indicated correctly. This paper describes the system and the operational procedures necessary to use the proposed method efficiently during a catheter intervention. Preliminary experimental results on a phantom are also reported.

Neocortical gray matter volume in first-episode schizophrenia and first-episode affective psychosis: a cross-sectional and longitudinal MRI study

BACKGROUND

Overall neocortical gray matter (NCGM) volume has not been studied in first-episode schizophrenia (FESZ) at first hospitalization or longitudinally to evaluate progression, nor has it been compared with first-episode affective psychosis (FEAFF).

METHODS

Expectation-maximization/atlas-based magnetic resonance imaging (MRI) tissue segmentation into gray matter, white matter (WM), or cerebrospinal fluid (CSF) at first hospitalization of 29 FESZ and 34 FEAFF, plus 36 matched healthy control subjects (HC), and, longitudinally approximately 1.5 years later, of 17 FESZ, 21 FEAFF, and 26 HC was done. Manual editing separated NCGM and its lobar parcellation, cerebral WM (CWM), lateral ventricles (LV), and sulcal CSF (SCSF).

RESULTS

At first hospitalization, FESZ and FEAFF showed smaller NCGM volumes and larger SCSF and LV than HC. Longitudinally, FESZ showed NCGM volume reduction (-1.7%), localized to frontal (-2.4%) and temporal (-2.6%) regions, and enlargement of SCSF (7.2%) and LV (10.4%). Poorer outcome was associated with these LV and NCGM changes. FEAFF showed longitudinal NCGM volume increases (3.6%) associated with lithium or valproate administration but without clinical correlations and regional localization.

CONCLUSIONS

Longitudinal NCGM volume reduction and CSF component enlargement in FESZ are compatible with post-onset progression. Longitudinal NCGM volume increase in FEAFF may reflect neurotrophic effects of mood stabilizers.

Automated segmentation of the actively stained mouse brain using multi-spectral MR microscopy

Magnetic resonance microscopy (MRM) has created new approaches for high-throughput morphological phenotyping of mouse models of diseases. Transgenic and knockout mice serve as a test bed for validating hypotheses that link genotype to the phenotype of diseases, as well as developing and tracking treatments. We describe here a Markov random fields based segmentation of the actively stained mouse brain, as a prerequisite for morphological phenotyping. Active staining achieves higher signal to noise ratio (SNR) thereby enabling higher resolution imaging per unit time than obtained in previous formalin-fixed mouse brain studies. The segmentation algorithm was trained on isotropic 43-mum T1- and T2-weighted MRM images. The mouse brain was segmented into 33 structures, including the hippocampus, amygdala, hypothalamus, thalamus, as well as fiber tracts and ventricles. Probabilistic information used in the segmentation consisted of (a) intensity distributions in the T1- and T2-weighted data, (b) location, and (c) contextual priors for incorporating spatial information. Validation using standard morphometric indices showed excellent consistency between automatically and manually segmented data. The algorithm has been tested on the widely used C57BL/6J strain, as well as on a selection of six recombinant inbred BXD strains, chosen especially for their largely variant hippocampus.

Multiclassifier fusion in human brain MR segmentation: modelling convergence

Segmentations of MR images of the human brain can be generated by propagating an existing atlas label volume to the target image. By fusing multiple propagated label volumes, the segmentation can be improved. We developed a model that predicts the improvement of labelling accuracy and precision based on the number of segmentations used as input. Using a cross-validation study on brain image data as well as numerical simulations, we verified the model. Fit parameters of this model are potential indicators of the quality of a given label propagation method or the consistency of the input segmentations used.

Automatic segmentation of the caudate nucleus from human brain MR images

We describe a knowledge-driven algorithm to automatically delineate the caudate nucleus (CN) region of the human brain from a magnetic resonance (MR) image. Since the lateral ventricles (LVs) are good landmarks for positioning the CN, the algorithm first extracts the LVs, and automatically localizes the CN from this information guided by anatomic knowledge of the structure. The face validity of the algorithm was tested with 55 high-resolution T1-weighted magnetic resonance imaging (MRI) datasets, and segmentation results were overlaid onto the original image data for visual inspection. We further evaluated the algorithm by comparing automated segmentation results to a "gold standard" established by human experts for these 55 MR datasets. Quantitative comparison showed a high intraclass correlation between the algorithm and expert as well as high spatial overlap between the regions-of-interest (ROIs) generated from the two methods. The mean spatial overlap +/- standard deviation (defined by the intersection of the 2 ROIs divided by the union of the 2 ROIs) was equal to 0.873 +/- 0.0234. The algorithm has been incorporated into a public domain software program written in Java and, thus, has the potential to be of broad benefit to neuroimaging investigators interested in basal ganglia anatomy and function.