Multiple sclerosis medical image analysis and information management

Handling changes in MRI acquisition parameters in modeling whole brain lesion volume and atrophy data in multiple sclerosis subjects: Comparison of linear mixed-effect models

Magnetic resonance imaging (MRI) of the brain provides important outcome measures in the longitudinal evaluation of disease activity and progression in MS subjects. Two common measures derived from brain MRI scans are the brain parenchymal fraction (BPF) and T2 hyperintense lesion volume (T2LV), and these measures are routinely assessed longitudinally in clinical trials and observational studies. When measuring each outcome longitudinally, observed changes may be potentially confounded by variability in MRI acquisition parameters between scans. In order to accurately model longitudinal change, the acquisition parameters should thus be considered in statistical models. In this paper, several models for including protocol as well as individual MRI acquisition parameters in linear mixed models were compared using a large dataset of 3453 longitudinal MRI scans from 1341 subjects enrolled in the CLIMB study, and model fit indices were compared across the models. The model that best explained the variance in BPF data was a random intercept and random slope with protocol specific residual variance along with the following fixed-effects: baseline age, baseline disease duration, protocol and study time. The model that best explained the variance in T2LV was a random intercept and random slope along with the following fixed-effects: baseline age, baseline disease duration, protocol and study time. In light of these findings, future studies pertaining to BPF and T2LV outcomes should carefully account for the protocol factors within longitudinal models to ensure that the disease trajectory of MS subjects can be assessed more accurately.

Using multiple imputation to efficiently correct cerebral MRI whole brain lesion and atrophy data in patients with multiple sclerosis

Automated segmentation of brain MRI scans into tissue classes is commonly used for the assessment of multiple sclerosis (MS). However, manual correction of the resulting brain tissue label maps by an expert reader remains necessary in many cases. Since automated segmentation data awaiting manual correction are "missing", we proposed to use multiple imputation (MI) to fill-in the missing manually-corrected MRI data for measures of normalized whole brain volume (brain parenchymal fraction-BPF) and T2 hyperintense lesion volume (T2LV). Automated and manually corrected MRI measures from 1300 patients enrolled in the Comprehensive Longitudinal Investigation of Multiple Sclerosis at the Brigham and Women's Hospital (CLIMB) were identified. Simulation studies were conducted to assess the performance of MI with missing data both missing completely at random and missing at random. An imputation model including the concurrent automated data as well as clinical and demographic variables explained a high proportion of the variance in the manually corrected BPF (R(2)=0.97) and T2LV (R(2)=0.89), demonstrating the potential to accurately impute the missing data. Further, our results demonstrate that MI allows for the accurate estimation of group differences with little to no bias and with similar precision compared to an analysis with no missing data. We believe that our findings provide important insights for efficient correction of automated MRI measures to obviate the need to perform manual correction on all cases.

HLA (A-B-C and -DRB1) alleles and brain MRI changes in multiple sclerosis: a longitudinal study

Several major histocompatibility complex (MHC) alleles have been postulated to influence the susceptibility to multiple sclerosis (MS), as well as its clinical/radiological course. In this longitudinal observation, we further explored the impact of human leukocyte antigen (HLA) class I/II alleles on MS outcomes, and we tested the hypothesis that HLA DRB1*1501 might uncover different strata of MS subjects harboring distinct MHC allele associations with magnetic resonance imaging (MRI) measures. Five hundred eighteen MS patients with two-digit HLA typing and at least one brain MRI were recruited for the study. T2-weighted hyperintense lesion volume (T2LV) and brain parenchymal fraction (BPF) were acquired at each time point. The association between allele count and MRI values was determined using linear regression modeling controlling for age, disease duration and gender. Analyses were also stratified by the presence/absence of HLA DRB1*1501. HLA DRB1*04 was associated with higher T2LV (P=0.006); after stratification, its significance remained only in the presence of HLA DRB1*1501 (P=0.012). The negative effect of HLA DRB1*14 on T2LV was exerted in DRB1*1501-negative group (P=0.012). Longitudinal analysis showed that HLA DRB1*10 was significantly protective on T2LV accrual in the presence of HLA DRB1*1501 (P=0.002). Although the majority of our results did not withstand multiple comparison correction, the differential impact of several HLA alleles in the presence/absence of HLA DRB1*1501 suggests that they may interact in determining the different phenotypic expressions of MS.

A putative Alzheimer's disease risk allele in PCK1 influences brain atrophy in multiple sclerosis

Background

Brain atrophy and cognitive dysfunction are neurodegenerative features of Multiple Sclerosis (MS). We used a candidate gene approach to address whether genetic variants implicated in susceptibility to late onset Alzheimer's Disease (AD) influence brain volume and cognition in MS patients.

Methods/Principal Findings

MS subjects were genotyped for five single nucleotide polymorphisms (SNPs) associated with susceptibility to AD: PICALM, CR1, CLU, PCK1, and ZNF224. We assessed brain volume using Brain Parenchymal Fraction (BPF) measurements obtained from Magnetic Resonance Imaging (MRI) data and cognitive function using the Symbol Digit Modalities Test (SDMT). Genotypes were correlated with cross-sectional BPF and SDMT scores using linear regression after adjusting for sex, age at symptom onset, and disease duration. 722 MS patients with a mean (±SD) age at enrollment of 41 (±10) years were followed for 44 (±28) months. The AD risk-associated allele of a non-synonymous SNP in the PCK1 locus (rs8192708^G) is associated with a smaller average brain volume (P = 0.0047) at the baseline MRI, but it does not impact our baseline estimate of cognition. PCK1 is additionally associated with higher baseline T2-hyperintense lesion volume (P = 0.0088). Finally, we provide technical validation of our observation in a subset of 641 subjects that have more than one MRI study, demonstrating the same association between PCK1 and smaller average brain volume (P = 0.0089) at the last MRI visit.

Conclusion/Significance

Our study provides suggestive evidence for greater brain atrophy in MS patients bearing the PCK1 allele associated with AD-susceptibility, yielding new insights into potentially shared neurodegenerative process between MS and late onset AD.

Neuroimaging study designs, computational analyses and data provenance using the LONI pipeline

Modern computational neuroscience employs diverse software tools and multidisciplinary expertise to analyze heterogeneous brain data. The classical problems of gathering meaningful data, fitting specific models, and discovering appropriate analysis and visualization tools give way to a new class of computational challenges—management of large and incongruous data, integration and interoperability of computational resources, and data provenance. We designed, implemented and validated a new paradigm for addressing these challenges in the neuroimaging field. Our solution is based on the LONI Pipeline environment [3], [4], a graphical workflow environment for constructing and executing complex data processing protocols. We developed study-design, database and visual language programming functionalities within the LONI Pipeline that enable the construction of complete, elaborate and robust graphical workflows for analyzing neuroimaging and other data. These workflows facilitate open sharing and communication of data and metadata, concrete processing protocols, result validation, and study replication among different investigators and research groups. The LONI Pipeline features include distributed grid-enabled infrastructure, virtualized execution environment, efficient integration, data provenance, validation and distribution of new computational tools, automated data format conversion, and an intuitive graphical user interface. We demonstrate the new LONI Pipeline features using large scale neuroimaging studies based on data from the International Consortium for Brain Mapping [5] and the Alzheimer's Disease Neuroimaging Initiative [6]. User guides, forums, instructions and downloads of the LONI Pipeline environment are available at http://pipeline.loni.ucla.edu.

HLA B*44: protective effects in MS susceptibility and MRI outcome measures

OBJECTIVE

In addition to the main multiple sclerosis (MS) major histocompatibility complex (MHC) risk allele (HLA DRB1*1501), investigations of the MHC have implicated several class I MHC loci (HLA A, HLA B, and HLA C) as potential independent MS susceptibility loci. Here, we evaluate the role of 3 putative protective alleles in MS: HLA A*02, HLA B*44, and HLA C*05.

METHODS

Subjects include a clinic-based patient sample with a diagnosis of either MS or a clinically isolated syndrome (n = 532), compared to subjects in a bone marrow donor registry (n = 776). All subjects have 2-digit HLA data. Logistic regression was used to determine the independence of each allele's effect. We used linear regression and an additive model to test for correlation between an allele and MRI and clinical measures of disease course.

RESULTS

After accounting for the effect of HLA DRB1*1501, both HLA A*02 and HLA B*44 are validated as susceptibility alleles (p(A*02) 0.00039 and p(B*44) 0.00092) and remain significantly associated with MS susceptibility in the presence of the other allele. Although A*02 is not associated with MS outcome measures, HLA B*44 demonstrates association with a better radiologic outcome both in terms of brain parenchymal fraction and T2 hyperintense lesion volume (p = 0.03 for each outcome).

CONCLUSION

The MHC class I alleles HLA A*02 and HLA B*44 independently reduce susceptibility to MS, but only HLA B*44 appears to influence disease course, preserving brain volume and reducing the burden of T2 hyperintense lesions in subjects with MS.

Is it time to re-prioritize neuroimaging databases and digital repositories?

The development of in vivo brain imaging has lead to the collection of large quantities of digital information. In any individual research article, several tens of gigabytes-worth of data may be represented-collected across normal and patient samples. With the ease of collecting such data, there is increased desire for brain imaging datasets to be openly shared through sophisticated databases. However, very often the raw and pre-processed versions of these data are not available to researchers outside of the team that collected them. A range of neuroimaging databasing approaches has streamlined the transmission, storage, and dissemination of data from such brain imaging studies. Though early sociological and technical concerns have been addressed, they have not been ameliorated altogether for many in the field. In this article, we review the progress made in neuroimaging databases, their role in data sharing, data management, potential for the construction of brain atlases, recording data provenance, and value for re-analysis, new publication, and training. We feature the LONI IDA as an example of an archive being used as a source for brain atlas workflow construction, list several instances of other successful uses of image databases, and comment on archive sustainability. Finally, we suggest that, given these developments, now is the time for the neuroimaging community to re-prioritize large-scale databases as a valuable component of brain imaging science.

Efficient, Distributed and Interactive Neuroimaging Data Analysis Using the LONI Pipeline

The LONI Pipeline is a graphical environment for construction, validation and execution of advanced neuroimaging data analysis protocols (Rex et al., 2003). It enables automated data format conversion, allows Grid utilization, facilitates data provenance, and provides a significant library of computational tools. There are two main advantages of the LONI Pipeline over other graphical analysis workflow architectures. It is built as a distributed Grid computing environment and permits efficient tool integration, protocol validation and broad resource distribution. To integrate existing data and computational tools within the LONI Pipeline environment, no modification of the resources themselves is required. The LONI Pipeline provides several types of process submissions based on the underlying server hardware infrastructure. Only workflow instructions and references to data, executable scripts and binary instructions are stored within the LONI Pipeline environment. This makes it portable, computationally efficient, distributed and independent of the individual binary processes involved in pipeline data-analysis workflows. We have expanded the LONI Pipeline (V.4.2) to include server-to-server (peer-to-peer) communication and a 3-tier failover infrastructure (Grid hardware, Sun Grid Engine/Distributed Resource Management Application API middleware, and the Pipeline server). Additionally, the LONI Pipeline provides three layers of background-server executions for all users/sites/systems. These new LONI Pipeline features facilitate resource-interoperability, decentralized computing, construction and validation of efficient and robust neuroimaging data-analysis workflows. Using brain imaging data from the Alzheimer's Disease Neuroimaging Initiative (Mueller et al., 2005), we demonstrate integration of disparate resources, graphical construction of complex neuroimaging analysis protocols and distributed parallel computing. The LONI Pipeline, its features, specifications, documentation and usage are available online (http://Pipeline.loni.ucla.edu).

Brain imaging of multiple sclerosis: the next 10 years

MR imaging has had a major impact on understanding the dynamic neuropathologic findings of multiple sclerosis (MS), early diagnosis of the disease, and clinical trial conduct. The next 10 years can be expected to see further advances with a greater emphasis on large multicenter studies, new techniques and hardware allowing greater imaging sensitivity and resolution, and the exploitation of positron emission tomography molecular imaging for MS. The impact should be felt with a new emphasis on gray matter disease and processes of repair. With new ways of monitoring the disease, new treatment targets should become practical, helping to translate advances in the understanding of immunology and regenerative medicine into novel therapies.

Provenance in neuroimaging

Provenance, the description of the history of a set of data, has grown more important with the proliferation of research consortia-related efforts in neuroimaging. Knowledge about the origin and history of an image is crucial for establishing data and results quality; detailed information about how it was processed, including the specific software routines and operating systems that were used, is necessary for proper interpretation, high fidelity replication and re-use. We have drafted a mechanism for describing provenance in a simple and easy to use environment, alleviating the burden of documentation from the user while still providing a rich description of an image's provenance. This combination of ease of use and highly descriptive metadata should greatly facilitate the collection of provenance and subsequent sharing of data.

Image registration framework for large-scale longitudinal MRI data sets: strategy and validation

Advanced magnetic resonance imaging (MRI) studies often require the transformation of large numbers of images into a common space. Calculating transformations that relate each image to every other and applying them to the images on demand are theoretically possible; however, these can be computationally prohibitive. Therefore, relating each image to only one other image, then linking those transforms together to relate any two images in the database, may be an efficient alternative. Evaluated were the feasibility and validity of image registration to bring intraindividual MR images into mutual correspondence for longitudinal analysis through the concatenation of precomputed transforms. A longitudinal data set of 10 multiple sclerosis patients with nine serial dual-echo spin-echo, 1.5-T MRI scans was used. Intrasubject registrations were performed stepwise between consecutive images and direct from each time point to the baseline. Consecutive transforms were concatenated and evaluated against direct registrations by comparing the resulting transformed images (using Pearson correlation coefficient). Confounding variables such as time between scans, brain atrophy, and change in lesion load were evaluated. We found the images resampled with the direct and the concatenated transforms to be highly correlated, and there was no significant difference between methods. Differences in brain parenchymal fraction (a measure of brain atrophy) showed significant inverse correlation with the correspondence of the resampled images. Results indicate that concatenating multiple transforms that link two images together produces near-identical results to that of direct registration; thus, this method is both useful and valid.

Web-PACS for Multicenter Clinical Trials

Medical information systems are not designed for clinical trials using clinical imaging. This paper presents a conceptual model for clinical trials based on medical imaging from two complementary points of view: a technical model and a business model. A Web information system (WIS) for supporting multicenter clinical trials has been designed to implement the proposed model. We show that our approach overcomes the actual limitations by facilitating medical image management in the context of clinical trials or cooperative research.

Magnetic resonance imaging advances in multiple sclerosis

Magnetic resonance imaging (MRI) has become a core component of clinical management and scientific research in multiple sclerosis (MS), providing essential information about tissue structure and function. MRI is now the most important laboratory diagnostic and longitudinal monitoring technology. A number of conventional MRI techniques, which include T2-weighted, T1-weighted, and gadolinium-enhanced imaging, are used to identify overt lesions and quantify tissue atrophy. MRI is highly sensitive in detecting brain and spinal cord involvement in MS and can visualize multifocal lesions, occult disease, and macroscopic atrophy. Advanced MRI techniques, such as magnetization transfer imaging, spectroscopy, diffusion-weighted imaging, and functional MRI, have added to our understanding of the pathogenesis of the disease. The precise role of these newer imaging approaches continues to be defined. In this supplement to the Journal of Neuroimaging, the authors review the role of conventional and advanced MRI techniques in detecting tissue changes in MS, diagnosing and monitoring patients, and charting the progression of disease in new and established patients.

Diffusion Tensor Magnetic Resonance Imaging in Multiple Sclerosis

Multiple sclerosis (MS), a demyelinating disease, occurs principally in the white matter (WM) of the central nervous system. Conventional magnetic resonance imaging (MRI) is sensitive to some, but not all, brain changes associated with MS. Diffusion-weighted imaging (DWI) provides information about water diffusion in tissue and diffusion tensor MRI (DT-MRI) about fiber direction, allowing for the identification of WM abnormalities that are not apparent on conventional MRI images. These techniques can quantitatively characterize the local microstructure of tissues. MS-associated disease processes lead to regions characterized by an increased amount of water diffusion and a decrease in the anisotropy of diffusion direction. These changes have been found to produce different patterns in MS patients presenting different courses of the disease. Changes in water diffusion may allow examination of the type, appearance, enhancement, and location of lesions not readily visible by other means. Ongoing studies of MS are integrating conventional MRI and DT-MRI measures with connectivity-based regional assessment, aiming to provide a better understanding of the nature and the location of WM lesions. This integration and the development of novel image-processing and visualization techniques may improve the understanding of WM architecture and its disruption in MS. This article presents a brief history of DWI, its basic principles and applications in the study of MS, a review of the properties and applications of DT-MRI, and their use in the study of MS. In addition, this article illustrates the methodology for the analysis of DT-MRI in ongoing studies of MS.

Magnetic resonance spectroscopy in the monitoring of multiple sclerosis

In addition to providing information on tissue structure, magnetic resonance (MR) technology offers the potential to investigate tissue metabolism and function. MR spectroscopy (MRS) offers a wealth of data on the biochemistry of a selected brain tissue volume, which represent potential surrogate markers for the pathology underlying multiple sclerosis (MS). In particular, the N-acetylaspartate peak in an MR spectrum is a putative marker of neuronal and axonal integrity, and the choline peak appears to reflect cell-membrane metabolism. On this basis, a diminished N-acetylaspartate peak is interpreted to represent neuronal/axonal dysfunction or loss, and an elevated choline peak represents heightened cell-membrane turnover, as seen in demyelination, remyelination, inflammation, or gliosis. Therefore, MRS may provide a unique tool to evaluate the severity of MS, establish a prognosis, follow disease evolution, understand its pathogenesis, and evaluate the efficacy of therapeutic interventions, which complements the information obtained from the various forms of assessment made by conventional MR imaging.