The use of atlas registration and graph cuts for prostate segmentation in magnetic resonance images

PURPOSE

An automatic method for 3D prostate segmentation in magnetic resonance (MR) images is presented for planning image-guided radiotherapy treatment of prostate cancer.

METHODS

A spatial prior based on intersubject atlas registration is combined with organ-specific intensity information in a graph cut segmentation framework. The segmentation is tested on 67 axial T2-weighted MR images in a leave-one-out cross validation experiment and compared with both manual reference segmentations and with multiatlas-based segmentations using majority voting atlas fusion. The impact of atlas selection is investigated in both the traditional atlas-based segmentation and the new graph cut method that combines atlas and intensity information in order to improve the segmentation accuracy. Best results were achieved using the method that combines intensity information, shape information, and atlas selection in the graph cut framework.

RESULTS

A mean Dice similarity coefficient (DSC) of 0.88 and a mean surface distance (MSD) of 1.45 mm with respect to the manual delineation were achieved.

CONCLUSIONS

This approaches the interobserver DSC of 0.90 and interobserver MSD 0f 1.15 mm and is comparable to other studies performing prostate segmentation in MR.

IT Infrastructure to support the secondary use of routinely acquired clinical imaging data for research

We propose an infrastructure for the automated anonymization, extraction and processing of image data stored in clinical data repositories to make routinely acquired imaging data available for research purposes. The automated system, which was tested in the context of analyzing routinely acquired MR brain imaging data, consists of four modules: subject selection using PACS query, anonymization of privacy sensitive information and removal of facial features, quality assurance on DICOM header and image information, and quantitative imaging biomarker extraction. In total, 1,616 examinations were selected based on the following MRI scanning protocols: dementia protocol (246), multiple sclerosis protocol (446) and open question protocol (924). We evaluated the effectiveness of the infrastructure in accessing and successfully extracting biomarkers from routinely acquired clinical imaging data. To examine the validity, we compared brain volumes between patient groups with positive and negative diagnosis, according to the patient reports. Overall, success rates of image data retrieval and automatic processing were 82.5 %, 82.3 % and 66.2 % for the three protocol groups respectively, indicating that a large percentage of routinely acquired clinical imaging data can be used for brain volumetry research, despite image heterogeneity. In line with the literature, brain volumes were found to be significantly smaller (p-value <0.001) in patients with a positive diagnosis of dementia (915 ml) compared to patients with a negative diagnosis (939 ml). This study demonstrates that quantitative image biomarkers such as intracranial and brain volume can be extracted from routinely acquired clinical imaging data. This enables secondary use of clinical images for research into quantitative biomarkers at a hitherto unprecedented scale.

Older age relates to worsening of fine motor skills: a population-based study of middle-aged and elderly persons

INTRODUCTION

In a population-based study of 1,912 community-dwelling persons of 45 years and older, we investigated the relation between age and fine motor skills using the Archimedes spiral-drawing test. Also, we studied the effect of brain volume on fine motor skills.

METHODS

Participants were required to trace a template of a spiral on an electronic drawing board. Clinical scores from this test were obtained by visual assessment of the drawings. Quantitative measures were objectively determined from the recorded data of the drawings. As tremor is known to occur increasingly with advancing age, we also rated drawings to assess presence of tremor.

RESULTS

We found presence of a tremor in 1.3% of the drawings. In the group without tremor, we found that older age was related to worse fine motor skills. Additionally, participants over the age of 75 showed increasing deviations from the template when drawing the spiral. Larger cerebral volume and smaller white matter lesion volume were related to better spiral-drawing performance, whereas cerebellar volume was not related to spiral-drawing performance.

CONCLUSION

Older age is related to worse fine motor skills, which can be captured by clinical scoring or quantitative measures of the Archimedes spiral-drawing test. Persons with a tremor performed worse on almost all measures of the spiral-drawing test. Furthermore, larger cerebral volume is related to better fine motor skills.

Hippocampal shape is predictive for the development of dementia in a normal, elderly population

Previous studies have shown that hippocampal volume is an early marker for dementia. We investigated whether hippocampal shape characteristics extracted from MRI scans are predictive for the development of dementia during follow up in subjects who were nondemented at baseline. Furthermore, we assessed whether hippocampal shape provides additional predictive value independent of hippocampal volume. Five hundred eleven brain MRI scans from elderly nondemented participants of a prospective population-based imaging study were used. During the 10-year follow-up period, 52 of these subjects developed dementia. For training and evaluation independent of age and gender, a subset of 50 cases and 150 matched controls was selected. The hippocampus was segmented using an automated method. From the segmentation, the volume was determined and a statistical shape model was constructed. We trained a classifier to distinguish between subjects who developed dementia and subjects who stayed cognitively healthy. For all subjects the a posteriori probability to develop dementia was estimated using the classifier in a cross-validation experiment. The area under the ROC curve for volume, shape, and the combination of both were, respectively, 0.724, 0.743, and 0.766. A logistic regression model showed that adding shape to a model using volume corrected for age and gender increased the global model-fit significantly (P = 0.0063). We conclude that hippocampal shape derived from MRI scans is predictive for dementia before clinical symptoms arise, independent of age and gender. Furthermore, the results suggest that hippocampal shape provides additional predictive value over hippocampal volume and that combining shape and volume leads to better prediction.

Tissue segmentation of head and neck CT images for treatment planning: a multiatlas approach combined with intensity modeling

PURPOSE

Hyperthermia treatment of head and neck tumors requires accurate treatment planning, based on 3D patient models that are derived from segmented 3D images. These segmentations are currently obtained by manual outlining of the relevant tissue regions, which is a tedious and time-consuming procedure (≈ 8 h) limiting the clinical applicability of hyperthermia treatment. In this context, the authors present and evaluate an automatic segmentation algorithm for CT images of the head and neck.

METHODS

The proposed method combines anatomical information, based on atlas registration, with local intensity information in a graph cut framework. The method is evaluated with respect to ground truth manual delineation and compared with multiatlas-based segmentation on a dataset of 18 labeled CT images using the Dice similarity coefficient (DSC), the mean surface distance (MSD), and the Hausdorff surface distance (HSD) as evaluation measures. On a subset of 13 labeled images, the influence of different labelers on the method's accuracy is quantified and compared with the interobserver variability.

RESULTS

For the DSC, the proposed method performs significantly better for the segmentation of all the tissues, except brain stem and spinal cord. The MSD shows a significant improvement for optical nerve, eye vitreous humor, lens, and thyroid. For the HSD, the proposed method performs significantly better for eye vitreous humor and brainstem. The proposed method has a significantly better score for DSC, MSD, and HSD than the multiatlas-based method for the eye vitreous humor. For the majority of the tissues (8/11) the segmentation accuracy of the proposed method is approaching the interobserver agreement. The authors' method showed better robustness to variations in atlas labeling compared with multiatlas segmentation. Moreover, the method improved the segmentation reproducibility compared with human observer's segmentations.

CONCLUSIONS

In conclusion, the proposed framework provides in an accurate automatic segmentation of head and neck tissues in CT images for the generation of 3D patient models, which improves reproducibility, and substantially reduces labor involved in therapy planning.

Total antioxidant capacity of the diet and major neurologic outcomes in older adults

OBJECTIVE

To evaluate total antioxidant capacity of the diet, measured by the ferric-reducing antioxidant power (FRAP) assay, in relation to risks of dementia and stroke, as well as key structural brain volumes, in the elderly.

METHODS

We prospectively studied 5,395 participants in the Rotterdam Study, aged 55 years and older, who were dementia free and provided dietary information at study baseline; 5,285 individuals were also stroke free at baseline, and 462 were dementia and stroke free at the time of an MRI brain scan 5 years after baseline. Dietary data were ascertained using a semiquantitative food-frequency questionnaire, and combined with food-specific FRAP measurements from published tables; this information was aggregated across the diet to obtain "dietary FRAP scores." Multivariable-adjusted Cox proportional hazard models were used to estimate relative risks of dementia and stroke, and multivariable-adjusted linear regression was used to estimate mean differences in structural brain volumes, across tertiles of dietary FRAP scores.

RESULTS

During a median 13.8 years of follow-up, we identified approximately 600 cases each of dementia and stroke. In multivariable-adjusted models, we observed no associations between dietary FRAP scores and risk of dementia (p trend = 0.3; relative risk = 1.12, 95% confidence interval = 0.91-1.38, comparing the highest vs lowest FRAP tertiles) or risk of stroke (p trend = 0.3; relative risk = 0.91, 95% confidence interval = 0.75-1.11, comparing extreme FRAP tertiles); results were similar across subtypes of these outcomes. Dietary FRAP scores were unrelated to brain tissue volumes as well.

CONCLUSIONS

Total antioxidant capacity of the diet, measured by dietary FRAP scores, does not seem to predict risks of major neurologic diseases.

The influence of cerebral small vessel disease on default mode network deactivation in mild cognitive impairment

Introduction

Cerebral small vessel disease (CSVD) is thought to contribute to cognitive dysfunction in patients with mild cognitive impairment (MCI). The underlying mechanisms, and more specifically, the effects of CSVD on brain functioning in MCI are incompletely understood. The objective of the present study was to examine the effects of CSVD on brain functioning, activation and deactivation, in patients with MCI using task-related functional MRI (fMRI).

Methods

We included 16 MCI patients with CSVD, 26 MCI patients without CSVD and 25 controls. All participants underwent a physical and neurological examination, neuropsychological testing, structural MRI, and fMRI during a graded working memory paradigm.

Results

MCI patients with and without CSVD had a similar neuropsychological profile and task performance during fMRI, but differed with respect to underlying (de)activation patterns. MCI patients with CSVD showed impaired deactivation in the precuneus/posterior cingulate cortex, a region known to be involved in the default mode network. In MCI patients without CSVD, brain activation depended on working memory load, as they showed relative ‘hyperactivation’ during vigilance, and ‘hypoactivation’ at a high working memory load condition in working memory related brain regions.

Conclusions

We present evidence that the potential underlying mechanism of CSVD affecting cognition in MCI is through network interference. The observed differences in brain activation and deactivation between MCI patients with and without CSVD, who had a similar ‘clinical phenotype’, support the view that, in patients with MCI, different types of pathology can contribute to cognitive impairment through different pathways.

The thyroid hormone receptor alpha locus and white matter lesions: a role for the clock gene REV-ERBα

BACKGROUND

Thyroid disorders are associated with an increased risk of cognitive impairment and Alzheimer's disease. Both small vessel disease and neurodegeneration have a role in the pathogenesis of cognitive impairment and Alzheimer's disease. Thyroid hormone receptor alpha (TRα) is the predominant TR in brain. The circadian clock gene REV-ERBα overlaps with the TRα gene and interferes with TRα expression. Limited data are available on the role of the TRα/REV-ERBα locus in small vessel disease and neurodegeneration. We therefore studied genetic variation in the TRα/REV-ERBα locus in relation to brain imaging data, as early markers for small vessel disease and neurodegeneration.

METHODS

Fifteen polymorphisms, covering the TRα/REV-ERBα locus, were studied in relation to white matter lesion (WML), total brain, and hippocampal volumes in the Rotterdam Study I (RS-I, n=454). Associations that remained significant after multiple testing correction were subsequently studied in an independent population for replication (RS-II, n=607).

RESULTS

No associations with total brain or hippocampal volumes were detected. A haplotype block in REV-ERBα was associated with WML volumes in RS-I. Absence of this haplotype was associated with larger WML volumes in women (0.38%±0.18% [β±SE], p=0.007), but not in men (0.04%±0.11%, p=0.24), which was replicated in RS-II (women: 0.15%±0.05%, p=0.04; men: 0.05%±0.07%, p=0.80). Meta-analysis of the two populations showed that women lacking this haplotype have a 1.9 times larger WML volume (p=0.001).

CONCLUSION

Our results suggest a role for REV-ERBα in the pathogenesis of WMLs.

A genome-wide association study identifies five loci influencing facial morphology in Europeans

Inter-individual variation in facial shape is one of the most noticeable phenotypes in humans, and it is clearly under genetic regulation; however, almost nothing is known about the genetic basis of normal human facial morphology. We therefore conducted a genome-wide association study for facial shape phenotypes in multiple discovery and replication cohorts, considering almost ten thousand individuals of European descent from several countries. Phenotyping of facial shape features was based on landmark data obtained from three-dimensional head magnetic resonance images (MRIs) and two-dimensional portrait images. We identified five independent genetic loci associated with different facial phenotypes, suggesting the involvement of five candidate genes--PRDM16, PAX3, TP63, C5orf50, and COL17A1--in the determination of the human face. Three of them have been implicated previously in vertebrate craniofacial development and disease, and the remaining two genes potentially represent novel players in the molecular networks governing facial development. Our finding at PAX3 influencing the position of the nasion replicates a recent GWAS of facial features. In addition to the reported GWA findings, we established links between common DNA variants previously associated with NSCL/P at 2p21, 8q24, 13q31, and 17q22 and normal facial-shape variations based on a candidate gene approach. Overall our study implies that DNA variants in genes essential for craniofacial development contribute with relatively small effect size to the spectrum of normal variation in human facial morphology. This observation has important consequences for future studies aiming to identify more genes involved in the human facial morphology, as well as for potential applications of DNA prediction of facial shape such as in future forensic applications.

Automated brain structure segmentation based on atlas registration and appearance models

Accurate automated brain structure segmentation methods facilitate the analysis of large-scale neuroimaging studies. This work describes a novel method for brain structure segmentation in magnetic resonance images that combines information about a structure's location and appearance. The spatial model is implemented by registering multiple atlas images to the target image and creating a spatial probability map. The structure's appearance is modeled by a classifier based on Gaussian scale-space features. These components are combined with a regularization term in a Bayesian framework that is globally optimized using graph cuts. The incorporation of the appearance model enables the method to segment structures with complex intensity distributions and increases its robustness against errors in the spatial model. The method is tested in cross-validation experiments on two datasets acquired with different magnetic resonance sequences, in which the hippocampus and cerebellum were segmented by an expert. Furthermore, the method is compared to two other segmentation techniques that were applied to the same data. Results show that the atlas- and appearance-based method produces accurate results with mean Dice similarity indices of 0.95 for the cerebellum, and 0.87 for the hippocampus. This was comparable to or better than the other methods, whereas the proposed technique is more widely applicable and robust.

Global and focal brain volume in long-term breast cancer survivors exposed to adjuvant chemotherapy

A limited number of studies have associated adjuvant chemotherapy with structural brain changes. These studies had small sample sizes and were conducted shortly after cessation of chemotherapy. Results of these studies indicate local gray matter volume decrease and an increase in white matter lesions. Up till now, it is unclear if non-CNS chemotherapy is associated with long-term structural brain changes. We compared focal and total brain volume (TBV) of a large set of non-CNS directed chemotherapy-exposed breast cancer survivors, on average 21 years post-treatment, to that of a population-based sample of women without a history of cancer. Structural MRI (1.5T) was performed in 184 chemotherapy-exposed breast cancer patients, mean age 64.0 (SD = 6.5) years, who had been diagnosed with cancer on average 21.1 (SD = 4.4) years before, and 368 age-matched cancer-free reference subjects from a population-based cohort study. Outcome measures were: TBV and total gray and white matter volume, and hippocampal volume. In addition, voxel based morphometry was performed to analyze differences in focal gray matter. The chemotherapy-exposed breast cancer survivors had significantly smaller TBV (-3.5 ml, P = 0.019) and gray matter volume (-2.9 ml, P = 0.003) than the reference subjects. No significant differences were observed in white matter volume, hippocampal volume, or local gray matter volume. This study shows that adjuvant chemotherapy for breast cancer is associated with long-term reductions in TBV and overall gray matter volume in the absence of focal reductions. The observed smaller gray matter volume in chemotherapy-exposed survivors was comparable to the effect of almost 4 years of age on gray matter volume reduction. These volume differences might be associated with the slightly worse cognitive performance that we observed previously in this group of breast cancer survivors.

Automated measurement of local white matter lesion volume

It has been hypothesized that white matter lesions at different locations may have different etiology and clinical consequences. Several approaches for the quantification of local white matter lesion load have been proposed in the literature, most of which rely on a distinction between lesions in a periventricular region close to the ventricles and a subcortical zone further away. In this work we present a novel automated method for local white matter lesion volume quantification in magnetic resonance images. The method segments and measures the white matter lesion volume in 43 regions defined by orientation and distance to the ventricles, which allows a more spatially detailed study of lesion load. The potential of the method was demonstrated by analyzing the effect of blood pressure on the regional white matter lesion volume in 490 elderly subjects taken from a longitudinal population study. The method was also compared to two commonly used techniques to assess the periventricular and subcortical lesion load. The main finding was that high blood pressure was primarily associated with lesion load in the vascular watershed area that forms the border between the periventricular and subcortical regions. It explains the associations found for both the periventricular and subcortical load computed for the same data, and that were reported in the literature. But the proposed method can localize the region of association with greater precision than techniques that distinguish between periventricular and subcortical lesions only.

A study of the bidirectional association between hippocampal volume on magnetic resonance imaging and depression in the elderly

BACKGROUND

Hippocampal volume loss on magnetic resonance imaging (MRI) has been reported in patients with depression. It is uncertain whether a small hippocampus renders a person vulnerable to develop depression or whether it is a consequence of depression. In this study, we addressed whether smaller baseline MRI hippocampal volumes increase the risk of incident depression. We also examined whether depressive symptoms at baseline were associated with decline in hippocampal volume during follow-up.

METHODS

Data were obtained in a prospective population-based study over a 10-year period. A sample of 514 nondemented persons aged 60 to 90 years underwent baseline measurements in 1995-1996 including three-dimensional MRI scans for assessment of hippocampal volumes and depressive symptoms (measured with Center for Epidemiologic Studies Depression Scale). Follow-up MRIs were made in 1999-2000 and in 2006. Incident depression was identified through standardized psychiatric examinations and continuous monitoring of medical and pharmaceutical records.

RESULTS

During a mean follow-up of 6.8 years per person (range .07-10.01 years), 135 of the 514 persons developed a clinically relevant episode of incident depressive symptoms. There was no association between baseline hippocampal volumes and incident depression (hazard ratio per SD decrease of average hippocampal volume .98 [.81-1.19], p = .84). A baseline Center for Epidemiologic Studies Depression Scale score of 16 or higher predicted a faster rate of decline in hippocampal volume. Also, incident depression was accompanied by a faster decline in left hippocampal volume.

CONCLUSIONS

This study provides no evidence that a small hippocampal volume precedes the development of late-life depression. Depression, however, may lead to a faster rate of hippocampal volume decline.

Statistical analysis of minimum cost path based structural brain connectivity

Diffusion MRI can be used to study the structural connectivity within the brain. Brain connectivity is often represented by a binary network whose topology can be studied using graph theory. We present a framework for the construction of weighted structural brain networks, containing information about connectivity, which can be effectively analyzed using statistical methods. Network nodes are defined by segmentation of subcortical structures and by cortical parcellation. Connectivity is established using a minimum cost path (mcp) method with an anisotropic local cost function based directly on diffusion weighted images. We refer to this framework as Statistical Analysis of Minimum cost path based Structural Connectivity (SAMSCo) and the weighted structural connectivity networks as mcp-networks. In a proof of principle study we investigated the information contained in mcp-networks by predicting subject age based on the mcp-networks of a group of 974 middle-aged and elderly subjects. Using SAMSCo, age was predicted with an average error of 3.7 years. This was significantly better than predictions based on fractional anisotropy or mean diffusivity averaged over the whole white matter or over the corpus callosum, which showed average prediction errors of at least 4.8 years. Additionally, we classified subjects, based on the mcp-networks, into groups with low and high white matter lesion load, while correcting for age, sex and white matter atrophy. The SAMSCo classification outperformed the classification based on the diffusion measures with a classification accuracy of 76.0% versus 63.2%. We also performed a classification in groups with mild and severe atrophy, correcting for age, sex and white matter lesion load. In this case, mcp-networks and diffusion measures yielded similar classification accuracies of 68.3% and 67.8% respectively. The SAMSCo prediction and classification experiments indicate that the mcp-networks contain information regarding age, white matter lesion load and white matter atrophy, and that in case of age and white matter lesion load the mcp-network based models outperformed the predictions based on diffusion measures.

A 10-year follow-up of hippocampal volume on magnetic resonance imaging in early dementia and cognitive decline

Hippocampal atrophy is frequently observed on magnetic resonance images from patients with Alzheimer's disease and persons with mild cognitive impairment. Even in asymptomatic elderly, a small hippocampal volume on magnetic resonance imaging is a risk factor for developing Alzheimer's disease. However, not everyone with a small hippocampus develops dementia. With the increased interest in the use of sequential magnetic resonance images as potential surrogate biomarkers of the disease process, it has also been shown that the rate of hippocampal atrophy is higher in persons with Alzheimer's disease compared to those with mild cognitive impairment and the healthy elderly. Whether a higher rate of hippocampal atrophy also predicts Alzheimer's disease or subtle cognitive decline in non-demented elderly is unknown. We examine these associations in a group of 518 elderly (age 60-90 years, 50% female), taken from the population-based Rotterdam Scan Study. A magnetic resonance imaging examination was performed at baseline in 1995-96 that was repeated in 1999-2000 (in 244 persons) and in 2006 (in 185 persons). Using automated segmentation procedures, we assessed hippocampal volumes on all magnetic resonance imaging scans. All persons were free of dementia at baseline and followed over time for cognitive decline and incident dementia. Persons had four repeated neuropsychological tests at the research centre over a 10-year period. We also continuously monitored the medical records of all 518 participants for incident dementia. During a total follow-up of 4360 person-years, (mean 8.4, range 0.1-11.3), 50 people developed incident dementia (36 had Alzheimer's disease). We found an increased risk to develop incident dementia per standard deviation faster rate of decline in hippocampal volume [left hippocampus 1.6 (95% confidence interval 1.2-2.3, right hippocampus 1.6 (95% confidence interval 1.2-2.1)]. Furthermore, decline in hippocampal volume predicted onset of clinical dementia when corrected for baseline hippocampal volume. In people who remained free of dementia during the whole follow-up period, we found that decline in hippocampal volume paralleled, and preceded, specific decline in delayed word recall. No associations were found in this sample between rate of hippocampal atrophy, Mini Mental State Examination and tests of executive function. Our results suggest that rate of hippocampal atrophy is an early marker of incipient memory decline and dementia, and could be of additional value when compared with a single hippocampal volume measurement as a surrogate biomarker of dementia.

Statistical analysis of structural brain connectivity

We present a framework for statistical analysis in large cohorts of structural brain connectivity, derived from diffusion weighted MRI. A brain network is defined between subcortical gray matter structures and a cortical parcellation obtained with FreeSurfer. Connectivity is established through minimum cost paths with an anisotropic local cost function and is quantified per connection. The connectivity network potentially encodes important information about brain structure, and can be analyzed using multivariate regression methods. The proposed framework can be used to study the relation between connectivity and e.g. brain function or neurodegenerative disease. As a proof of principle, we perform principal component regression in order to predict age and gender, based on the connectivity networks of 979 middle-aged and elderly subjects, in a 10-fold cross-validation. The results are compared to predictions based on fractional anisotropy and mean diffusivity averaged over the white matter and over the corpus callosum. Additionally, the predictions are performed based on the best predicting connection in the network. Principal component regression outperformed all other prediction models, demonstrating the age and gender information encoded in the connectivity network.

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.

Brain tissue volumes in the general elderly population

We investigated how volumes of cerebrospinal fluid (CSF), grey matter (GM) and white matter (WM) varied with age, sex, small vessel disease and cardiovascular risk factors in the Rotterdam Scan Study. Participants (n=490; 60-90 years) were non-demented and 51.0% had hypertension, 4.9% had diabetes mellitus, 17.8% were current smoker and 54.0% were former smoker. We segmented brain MR-images into GM, normal WM, white matter lesion (WML) and CSF. Brain infarcts were rated visually. Volumes were expressed as percentage of intra-cranial volume. With increasing age, volumes of total brain, normal WM and total WM decreased; that of GM remained unchanged; and that of WML increased, in both men and women. Excluding persons with infarcts did not alter these results. Persons with larger load of small vessel disease had smaller brain volume, especially normal WM volume. Diastolic blood pressure, diabetes mellitus and current smoking were also related to smaller brain volume. In the elderly, higher age, small vessel disease and cardiovascular risk factors are associated with smaller brain volume, especially WM volume.

Multi-spectral brain tissue segmentation using automatically trained k-Nearest-Neighbor classification

Conventional k-Nearest-Neighbor (kNN) classification, which has been successfully applied to classify brain tissue in MR data, requires training on manually labeled subjects. This manual labeling is a laborious and time-consuming procedure. In this work, a new fully automated brain tissue classification procedure is presented, in which kNN training is automated. This is achieved by non-rigidly registering the MR data with a tissue probability atlas to automatically select training samples, followed by a post-processing step to keep the most reliable samples. The accuracy of the new method was compared to rigid registration-based training and to conventional kNN-based segmentation using training on manually labeled subjects for segmenting gray matter (GM), white matter (WM) and cerebrospinal fluid (CSF) in 12 data sets. Furthermore, for all classification methods, the performance was assessed when varying the free parameters. Finally, the robustness of the fully automated procedure was evaluated on 59 subjects. The automated training method using non-rigid registration with a tissue probability atlas was significantly more accurate than rigid registration. For both automated training using non-rigid registration and for the manually trained kNN classifier, the difference with the manual labeling by observers was not significantly larger than inter-observer variability for all tissue types. From the robustness study, it was clear that, given an appropriate brain atlas and optimal parameters, our new fully automated, non-rigid registration-based method gives accurate and robust segmentation results. A similarity index was used for comparison with manually trained kNN. The similarity indices were 0.93, 0.92 and 0.92, for CSF, GM and WM, respectively. It can be concluded that our fully automated method using non-rigid registration may replace manual segmentation, and thus that automated brain tissue segmentation without laborious manual training is feasible.