Triangulating brain alterations in anorexia nervosa: a multimodal investigation of magnetic resonance spectroscopy, morphometry and blood-based biomarkers.

The acute state of anorexia nervosa (AN) is associated with widespread reductions in cortical gray matter (GM) thickness and white matter (WM) volume, suspected changes in myelin content and elevated levels of the neuronal damage marker neurofilament light (NF-L), but the underlying mechanisms remain largely unclear. To gain a deeper understanding of brain changes in AN, we applied a multimodal approach combining advanced neuroimaging methods with analysis of blood-derived biomarkers. In addition to standard measures of cortical GM thickness and WM volume, we analyzed tissue-specific profiles of brain metabolites using multivoxel proton magnetic resonance spectroscopy, T1 relaxation time as a proxy of myelin content leveraging advanced quantitative MRI methods and serum NF-L concentrations in a sample of 30 female, predominately adolescent patients with AN and 30 age-matched female healthy control participants. In patients with AN, we found a reduction in GM cortical thickness and GM total N-acetyl aspartate. The latter predicted higher NF-L levels, which were elevated in AN. Furthermore, GM total choline was elevated. In WM, there were no group differences in either imaging markers, choline levels or N-acetyl aspartate levels. The current study provides evidence for neuronal damage processes as well as for increased membrane lipid catabolism and turnover in GM in acute AN but no evidence for WM pathology. Our results illustrate the potential of multimodal research including tissue-specific proton magnetic resonance spectroscopy analyses to shed light on brain changes in psychiatric and neurological conditions, which may ultimately lead to better treatments.

Unchanged perfusion in normal-appearing white and grey matter of glioma patients nine months after proton beam irradiation.

PURPOSE: Radio(chemo)therapy is used as a standard treatment for glioma patients. The surrounding normal tissue is inevitably affected by the irradiation. The aim of this longitudinal study was to investigate perfusion alterations in the normal-appearing tissue after proton irradiation and assess the dose sensitivity of the normal tissue perfusion. METHODS: In 14 glioma patients, a sub-cohort of a prospective clinical trial (NCT02824731), perfusion changes in normal-appearing white matter (WM), grey matter (GM) and subcortical GM structures, i.e. caudate nucleus, hippocampus, amygdala, putamen, pallidum and thalamus, were evaluated before treatment and at three-monthly intervals after proton beam irradiation. The relative cerebral blood volume (rCBV) was assessed with dynamic susceptibility contrast MRI and analysed as the percentage ratio between follow-up and baseline image (ΔrCBV). Radiation-induced alterations were evaluated using Wilcoxon signed rank test. Dose and time correlations were investigated with univariate and multivariate linear regression models. RESULTS: No significant ΔrCBV changes were found in any normal-appearing WM and GM region after proton beam irradiation. A positive correlation with radiation dose was observed in the multivariate regression model applied to the combined ΔrCBV values of low (1-20 Gy), intermediate (21-40 Gy) and high (41-60 Gy) dose regions of GM (p < 0.001), while no time dependency was detected in any normal-appearing area. CONCLUSION: The perfusion in normal-appearing brain tissue remained unaltered after proton beam therapy. In further studies, a direct comparison with changes after photon therapy is recommended to confirm the different effect of proton therapy on the normal-appearing tissue.

Increased relative biological effectiveness and periventricular radiosensitivity in proton therapy of glioma patients.

PURPOSE: Currently, there is an intense debate on variations in intra-cerebral radiosensitivity and relative biological effectiveness (RBE) in proton therapy of primary brain tumours. Here, both effects were retrospectively investigated using late radiation-induced brain injuries (RIBI) observed in follow-up after proton therapy of patients with diagnosed glioma. METHODS: In total, 42 WHO grade 2-3 glioma patients out of a consecutive patient cohort having received (adjuvant) proton radio(chemo)therapy between 2014 and 2017 were eligible for analysis. RIBI lesions (symptomatic or clinically asymptomatic) were diagnosed and delineated on contrast-enhanced T1-weighted magnetic resonance imaging scans obtained in the first two years of follow-up. Correlation of RIBI location and occurrence with dose (D), proton dose-averaged linear energy transfer (LET) and variable RBE dose parameters were tested in voxel- and in patient-wise logistic regression analyses. Additionally, anatomical and clinical parameters were considered. Model performance was estimated through cross-validated area-under-the-curve (AUC) values. RESULTS: In total, 64 RIBI lesions were diagnosed in 21 patients. The median time between start of proton radio(chemo)therapy and RIBI appearance was 10.2 months. Median distances of the RIBI volume centres to the cerebral ventricles and to the clinical target volume border were 2.1 mm and 1.3 mm, respectively. In voxel-wise regression, the multivariable model with D, D × LET and periventricular region (PVR) revealed the highest AUC of 0.90 (95 % confidence interval: 0.89-0.91) while the corresponding model without D × LET revealed a value of 0.84 (0.83-0.86). In patient-level analysis, the equivalent uniform dose (EUD11, a = 11) in the PVR using a variable RBE was the most prominent predictor for RIBI with an AUC of 0.63 (0.32-0.90). CONCLUSIONS: In this glioma cohort, an increased radiosensitivity within the PVR was observed as well as a spatial correlation of RIBI with an increased RBE. Both need to be considered when delivering radio(chemo)therapy using proton beams.

Subjective memory impairment in glioma patients with curative radiotherapy.

BACKGROUND: Radiotherapy in patients with primary brain tumors may affect hippocampal structure and cause dyscognitive side-effects. PATIENTS AND METHODS: Using structural MRI and comprehensive neurocognitive evaluation, we investigated associations between hippocampal structure and memory deficits in 15 patients with WHO grade 3 and grade 4 gliomas receiving standard radio(chemo)therapy. RESULTS: We did not find changes in hippocampal thickness or cognitive abilities three months after completing radiotherapy. However, subjective memory impairment was associated with symptoms of depression, but not with objective memory performance, cortical thickness of the hippocampus or radiation dose. CONCLUSIONS: Irrespective of whether there is a bidirectional relationship between affective changes and subjective cognitive dysfunction in these patients, depressive symptoms remain a target for intervention to improve their quality of life. The results of our pilot study highlight that future assessment of side effects of radiotherapy concerning memory should include assessments of depressive symptoms.

MR Image Changes of Normal-Appearing Brain Tissue after Radiotherapy.

Radiotherapy is part of the standard treatment of most primary brain tumors. Large clinical target volumes and physical characteristics of photon beams inevitably lead to irradiation of surrounding normal brain tissue. This can cause radiation-induced brain injury. In particular, late brain injury, such as cognitive dysfunction, is often irreversible and progressive over time, resulting in a significant reduction in quality of life. Since 50% of patients have survival times greater than six months, radiation-induced side effects become more relevant and need to be balanced against radiation treatment given with curative intent. To develop adequate treatment and prevention strategies, the underlying cause of radiation-induced side-effects needs to be understood. This paper provides an overview of radiation-induced changes observed in normal-appearing brains measured with conventional and advanced MRI techniques and summarizes the current findings and conclusions. Brain atrophy was observed with anatomical MRI. Changes in tissue microstructure were seen on diffusion imaging. Vascular changes were examined with perfusion-weighted imaging and susceptibility-weighted imaging. MR spectroscopy revealed decreasing N-acetyl aspartate, indicating decreased neuronal health or neuronal loss. Based on these findings, multicenter prospective studies incorporating advanced MR techniques as well as neurocognitive function tests should be designed in order to gain more evidence on radiation-induced sequelae.

Dose dependent cerebellar atrophy in glioma patients after radio(chemo)therapy.

BACKGROUND AND PURPOSE: Radiotherapy is a standard treatment option for high-grade gliomas. Brain atrophy has previously been associated with radiotherapy. The goal of this study was to investigate dose dependent cerebellar atrophy using prospective, longitudinal MR data from adult glioma patients who received radiotherapy. MATERIALS AND METHODS: Cerebellar volumes were measured using T1-weighted MR images from 91 glioma patients before radiotherapy (N = 91) and from longitudinal follow-ups acquired in three monthly intervals (N = 349). Relative cerebellar volumes were calculated as ratios to the corresponding baseline values. Univariate mixed effects models were used to determine factors that were significantly associated with relative cerebellar volumes. These factors were subsequently included as fixed effects in a final multivariate linear mixed effects model. RESULTS: In multivariate analysis, cerebellar volume decreased significantly as a function of time (p < 0.001), time × dose (p < 0.001) and patient age (p = 0.007). Considering a 55 year patient receiving a mean cerebellar dose of 0 Gy (10 Gy), the linear mixed effects model predicts a relative cerebellar volume loss of 0.4% (2.0%) after 1 year and 0.7% (3.6%) after 2 years. Compared to patients treated with photons, the cerebellar dose was significantly lower in patients treated with proton therapy (p < 0.001, r = 0.62). CONCLUSION: Cerebellar volume decreased significantly and irreversibly after radiotherapy as function of time and mean cerebellar dose. Further work is now needed to correlate these results with cognitive function and motor performance.

Late Side Effects in Normal Mouse Brain Tissue After Proton Irradiation.

Radiation-induced late side effects such as cognitive decline and normal tissue complications can severely affect quality of life and outcome in long-term survivors of brain tumors. Proton therapy offers a favorable depth-dose deposition with the potential to spare tumor-surrounding normal tissue, thus potentially reducing such side effects. In this study, we describe a preclinical model to reveal underlying biological mechanisms caused by precise high-dose proton irradiation of a brain subvolume. We studied the dose- and time-dependent radiation response of mouse brain tissue, using a high-precision image-guided proton irradiation setup for small animals established at the University Proton Therapy Dresden (UPTD). The right hippocampal area of ten C57BL/6 and ten C3H/He mice was irradiated. Both strains contained four groups (nirradiated = 3, ncontrol = 1) treated with increasing doses (0 Gy, 45 Gy, 65 Gy or 85 Gy and 0 Gy, 40 Gy, 60 Gy or 80 Gy, respectively). Follow-up examinations were performed for up to six months, including longitudinal monitoring of general health status and regular contrast-enhanced magnetic resonance imaging (MRI) of mouse brains. These findings were related to comprehensive histological analysis. In all mice of the highest dose group, first symptoms of blood-brain barrier (BBB) damage appeared one week after irradiation, while a dose-dependent delay in onset was observed for lower doses. MRI contrast agent leakage occurred in the irradiated brain areas and was progressive in the higher dose groups. Mouse health status and survival corresponded to the extent of contrast agent leakage. Histological analysis revealed tissue changes such as vessel abnormalities, gliosis, and granule cell dispersion, which also partly affected the non-irradiated contralateral hippocampus in the higher dose groups. All observed effects depended strongly on the prescribed radiation dose and the outcome, i.e. survival, image changes, and tissue alterations, were very consistent within an experimental dose cohort. The derived dose-response model will determine endpoint-specific dose levels for future experiments and may support generating clinical hypotheses on brain toxicity after proton therapy.

Multiparametric cerebellar imaging and clinical phenotype in childhood ataxia telangiectasia.

BACKGROUND: Ataxia Telangiectasia (A-T) is an inherited multisystem disorder with cerebellar neurodegeneration. The relationships between imaging metrics of cerebellar health and neurological function across childhood in A-T are unknown, but may be important for determining timing and impact of therapeutic interventions. PURPOSE: To test the hypothesis that abnormalities of cerebellar structure, physiology and cellular health occur in childhood A-T and correlate with neurological disability, we performed multiparametric cerebellar MRI and establish associations with disease status in childhood A-T. METHODS: Prospective cross-sectional observational study. 22 young people (9 females / 13 males, age 6.6-17.8 years) with A-T and 24 matched healthy controls underwent 3-Tesla MRI with volumetric, diffusion and proton spectroscopic acquisitions. Participants with A-T underwent structured neurological assessment, and expression / activity of ataxia-telangiectasia mutated (ATM) kinase were recorded. RESULTS: Ataxia-telangiectasia participants had cerebellar volume loss (fractional total cerebellar volume: 5.3% vs 8.7%, P < 0.0005, fractional 4th ventricular volumes: 0.19% vs 0.13%, P < 0.0005), that progressed with age (fractional cerebellar volumes, r = -0.66, P = 0.001), different from the control group (t = -4.88, P < 0.0005). The relationship between cerebellar volume and age was similar for A-T participants with absent ATM kinase production and those producing non-functioning ATM kinase. Markers of cerebellar white matter injury were elevated in ataxia-telangiectasia vs controls (apparent diffusion coefficient: 0.89 × 10-3 mm2 s-1 vs 0.69 × 10-3 mm2 s-1, p < 0.0005) and correlated (age-corrected) with neurometabolite ratios indicating impaired neuronal viability (N-acetylaspartate:creatine r = -0.70, P < 0.001); gliosis (inositol:creatine r = 0.50, P = 0.018; combined glutamine/glutamate:creatine r = -0.55, P = 0.008) and increased myelin turnover (choline:creatine r = 0.68, P < 0.001). Fractional 4th ventricular volume was the only variable retained in the regression model predicting neurological function (adjusted r2 = 0.29, P = 0.015). CONCLUSIONS: Quantitative MRI demonstrates cerebellar abnormalities in children with A-T, providing non-invasive measures of progressive cerebellar injury and markers reflecting neurological status. These MRI metrics may be of value in determining timing and impact of interventions aimed at altering the natural history of A-T.

Tissue-type mapping of gliomas.

PURPOSE: To develop a statistical method of combining multimodal MRI (mMRI) of adult glial brain tumours to generate tissue heterogeneity maps that indicate tumour grade and infiltration margins. MATERIALS AND METHODS: We performed a retrospective analysis of mMRI from patients with histological diagnosis of glioma (n = 25). 1H Magnetic Resonance Spectroscopic Imaging (MRSI) was used to label regions of "pure" low- or high-grade tumour across image types. Normal brain and oedema characteristics were defined from healthy controls (n = 10) and brain metastasis patients (n = 10) respectively. Probability density distributions (PDD) for each tissue type were extracted from intensity normalised proton density and T2-weighted images, and p and q diffusion maps. Superpixel segmentation and Bayesian inference was used to produce whole-brain tissue-type maps. RESULTS: Total lesion volumes derived automatically from tissue-type maps correlated with those from manual delineation (p < 0.001, r = 0.87). Large high-grade volumes were determined in all grade III & IV (n = 16) tumours, in grade II gemistocytic rich astrocytomas (n = 3) and one astrocytoma with a histological diagnosis of grade II. For patients with known outcome (n = 20), patients with survival time < 2 years (3 grade II, 2 grade III and 10 grade IV) had a high-grade volume significantly greater than zero (Wilcoxon signed rank p < 0.0001) and also significantly greater high grade volume than the 5 grade II patients with survival >2 years (Mann Witney p = 0.0001). Regions classified from mMRI as oedema had non-tumour-like 1H MRS characteristics. CONCLUSIONS: 1H MRSI can label tumour tissue types to enable development of a mMRI tissue type mapping algorithm, with potential to aid management of patients with glial tumours.

Retrospective assessment of MRI-based volumetric changes of normal tissues in glioma patients following radio(chemo)therapy.

In glioma patients, linac-based photon beam irradiation is a widely applied therapy, which achieves highly conformal target volume coverage, but is also known to cause side-effects to adjacent areas of healthy tissue. Apart from subjective measures, such as quality of life assessment and neurocognitive function tests, objective methods to quantify tissue damage are needed to assess this impact. Magnetic resonance imaging (MRI) is a well-established method for brain tumor diagnoses as well as assessing treatment response. In this study, we retrospectively assessed volumetric changes of gray matter (GM) and white matter (WM) in glioma patients following photon irradiation using a heterogeneous MRI-dataset obtained in routine clinical practice at different sites with imaging parameters and magnetic field strengths. We found a significant reduction in WM volume at one year [Formula: see text] and two years [Formula: see text] post radio(chemo)therapy whereas corresponding GM volumes did not change significantly ([Formula: see text] and [Formula: see text], respectively). More importantly, we also found large variations in the segmented tissue volumes caused by the heterogeneous MR data, thus potentially masking more subtle tissue changes over time. On the basis of these observations, we present suggestions regarding data acquisitions in future prospective MR studies to assess such volumetric changes.

Targeted transcranial theta-burst stimulation alters fronto-insular network and prefrontal GABA.

Repetitive transcranial magnetic stimulation (rTMS) has been used worldwide to treat depression. However, the exact physiological effects are not well understood. Pathophysiology of depression involves crucial limbic structures (e.g. insula), and it is still not clear if these structures can be modulated through neurostimulation of surface regions (e.g. dorsolateral prefrontal cortex, DLPFC), and whether rTMS-induced excitatory/inhibitory transmission alterations relate to fronto-limbic connectivity changes. Therefore, we sought proof-of-concept for neuromodulation of insula via prefrontal intermittent theta-burst stimulation (iTBS), and how these effects relate to GABAergic and glutamatergic systems. In 27 healthy controls, we employed a single-blind crossover randomised-controlled trial comparing placebo and real iTBS using resting-state functional MRI and magnetic resonance spectroscopy. Granger causal analysis was seeded from right anterior insula (rAI) to locate individualized left DLPFC rTMS targets. Effective connectivity coefficients within rAI and DLPFC were calculated, and levels of GABA/Glx, GABA/Cr and Glx/Cr in DLPFC and anterior cingulate voxels were also measured. ITBS significantly dampened fronto-insular connectivity and reduced GABA/Glx in both voxels. GABA/Glx had a significant mediating effect on iTBS-induced changes in DLPFC-to-rAI connectivity. We demonstrate modulation of the rAI using targeted iTBS through alterations of excitatory/inhibitory interactions, which may underlie therapeutic effects of rTMS, offering promise for rTMS treatment optimization.

Cingulate GABA levels inversely correlate with the intensity of ongoing chronic knee osteoarthritis pain.

BACKGROUND: This study aims to investigate the role of the mid-anterior cingulate cortex γ-aminobutyric acid levels in chronic nociceptive pain. The molecular mechanisms of pain chronification are not well understood. In fibromyalgia, low mid-anterior cingulate cortex γ-aminobutyric acid was associated with high pain suggesting a role of prefrontal disinhibition. We hypothesize that mid-anterior cingulate cortex GABAergic disinhibition may underpin chronic pain independent of the pain etiology and comorbid negative affect. Proton magnetic resonance spectra were acquired at 3T from the mid-anterior cingulate cortex in 20 patients with chronic painful knee osteoarthritis, and 19 healthy pain-free individuals using a point resolved spectroscopy sequence optimized for detection of γ-aminobutyric acid. Participants underwent questionnaires for negative affect (depression and anxiety) and psychophysical pain phenotyping. RESULTS: No differences in mid-anterior cingulate cortex γ-aminobutyric acid or other metabolite levels were detected between groups. Ratings of perceived intensity of ongoing osteoarthritis pain were inversely correlated with γ-aminobutyric acid (r = -0.758, p < 0.001), but no correlations were seen for negative affect or pain thresholds. The pain γ-aminobutyric acid interrelation remained strong when controlling for depression (r = -0.820, p < 0.001). Combined levels of glutamine and glutamate were unrelated to psychometric or to pain thresholds. CONCLUSION: Our study supports mid-anterior cingulate cortex γ-aminobutyric acid as a potential marker of pain severity in chronic nociceptive pain states independent of negative affect. The findings suggest that GABAergic disinhibition of the salience network may underlie sensitization to averse stimuli as a mechanism contributing to pain chronification.

Multi-centre reproducibility of diffusion MRI parameters for clinical sequences in the brain.

The purpose of this work was to assess the reproducibility of diffusion imaging, and in particular the apparent diffusion coefficient (ADC), intra-voxel incoherent motion (IVIM) parameters and diffusion tensor imaging (DTI) parameters, across multiple centres using clinically available protocols with limited harmonization between sequences. An ice-water phantom and nine healthy volunteers were scanned across fives centres on eight scanners (four Siemens 1.5T, four Philips 3T). The mean ADC, IVIM parameters (diffusion coefficient D and perfusion fraction f) and DTI parameters (mean diffusivity MD and fractional anisotropy FA), were measured in grey matter, white matter and specific brain sub-regions. A mixed effect model was used to measure the intra- and inter-scanner coefficient of variation (CV) for each of the five parameters. ADC, D, MD and FA had a good intra- and inter-scanner reproducibility in both grey and white matter, with a CV ranging between 1% and 7.4%; mean 2.6%. Other brain regions also showed high levels of reproducibility except for small structures such as the choroid plexus. The IVIM parameter f had a higher intra-scanner CV of 8.4% and inter-scanner CV of 24.8%. No major difference in the inter-scanner CV for ADC, D, MD and FA was observed when analysing the 1.5T and 3T scanners separately. ADC, D, MD and FA all showed good intra-scanner reproducibility, with the inter-scanner reproducibility being comparable or faring slightly worse, suggesting that using data from multiple scanners does not have an adverse effect compared with using data from the same scanner. The IVIM parameter f had a poorer inter-scanner CV when scanners of different field strengths were combined, and the parameter was also affected by the scan acquisition resolution. This study shows that the majority of diffusion MRI derived parameters are robust across 1.5T and 3T scanners and suitable for use in multi-centre clinical studies and trials.

Manifold Learning in MR spectroscopy using nonlinear dimensionality reduction and unsupervised clustering.

PURPOSE: To investigate whether nonlinear dimensionality reduction improves unsupervised classification of (1) H MRS brain tumor data compared with a linear method. METHODS: In vivo single-voxel (1) H magnetic resonance spectroscopy (55 patients) and (1) H magnetic resonance spectroscopy imaging (MRSI) (29 patients) data were acquired from histopathologically diagnosed gliomas. Data reduction using Laplacian eigenmaps (LE) or independent component analysis (ICA) was followed by k-means clustering or agglomerative hierarchical clustering (AHC) for unsupervised learning to assess tumor grade and for tissue type segmentation of MRSI data. RESULTS: An accuracy of 93% in classification of glioma grade II and grade IV, with 100% accuracy in distinguishing tumor and normal spectra, was obtained by LE with unsupervised clustering, but not with the combination of k-means and ICA. With (1) H MRSI data, LE provided a more linear distribution of data for cluster analysis and better cluster stability than ICA. LE combined with k-means or AHC provided 91% accuracy for classifying tumor grade and 100% accuracy for identifying normal tissue voxels. Color-coded visualization of normal brain, tumor core, and infiltration regions was achieved with LE combined with AHC. CONCLUSION: The LE method is promising for unsupervised clustering to separate brain and tumor tissue with automated color-coding for visualization of (1) H MRSI data after cluster analysis.

(1) H 2D MRSI tissue type analysis of gliomas.

PURPOSE: To decompose 1H MR spectra of glioma patients into normal and abnormal tissue proportions for tumor classification and delineation. METHODS: Anatomical imaging and 1H magnetic resonance spectroscopic imaging data have been acquired from 11 grade II and 13 grade IV glioma patients. LCModel was used to decompose the magnetic resonance spectroscopic imaging data into normal brain, grade II, and grade IV tissue proportions using a tissue type basis set. Simulations were conducted to evaluate the accuracy of the methodology. Results were visualized using colormaps and abnormality contours showing tumor grade and extent. RESULTS: Simulations suggest that infiltrative tumor proportions as low as 20% can be identified at the typical 1H magnetic resonance spectroscopy signal-to-noise found in vivo. Tumor grading according to the highest estimated tumor grade within a lesion gave a classification accuracy of 86% discriminating between grade II and grade IV glioma. Voxels with significant proportions of tumor type spectra were found beyond the margins of contrast enhancement for most grade IV cases consistent with infiltration whereas the abnormality contours show that some tumors are confined within the hyperintensities shown by both post contrast T1 weighted and T2 weighted imaging. CONCLUSION: LCModel can be used to decompose 1H MR spectra into proportions of normal and abnormal tissue to identify tumor extent, infiltration, and overall grade.

Classification of single-voxel 1H spectra of childhood cerebellar tumors using LCModel and whole tissue representations.

In this study, mean tumor spectra are used as the basis functions in LCModel to create a direct classification tool for short echo time (1)H magnetic resonance spectroscopy of pediatric brain tumors. LCModel is a widely used analysis tool designed to fit a linear combination of individual metabolite spectra to in vivo spectra. Here, we have used LCModel to fit mean spectra and corresponding variability components of childhood cerebellar tumors, as calculated using principal component analysis, and assessed for classification accuracy. Classification was performed according to the highest estimated tumor proportion. This method was tested in a leave-one-out analysis discriminating between pediatric brain tumor spectra of medulloblastoma vs. pilocytic astrocytoma and medulloblastoma vs. pilocytic astrocytoma vs. ependymoma. Additionally, the effect of accepting different Cramér-Rao Lower Bound cut-off criteria on classification accuracy and estimated tissue proportions was investigated. The best classification results differentiating medulloblastoma vs. pilocytic astrocytoma and medulloblastoma vs. pilocytic astrocytoma vs. ependymoma were 100 and 87.7%, respectively. These results are comparable to a specialized pattern recognition analysis of this data set and give easy to interpret results in the form of estimated tissue proportions. The method requires minimal user input and is easily transferable across sites and to other magnetic resonance spectroscopy classification problems.