Towards Longitudinal Glioma Segmentation: Evaluating combined pre- and post-treatment MRI training data for automated tumor segmentation using nnU-Net.

Identification of key phenotypic regions such as necrosis, contrast enhancement, and edema on magnetic resonance imaging (MRI) is important for understanding disease evolution and treatment response in patients with glioma. Manual delineation is time intensive and not feasible for a clinical workflow. Automating phenotypic region segmentation overcomes many issues with manual segmentation, however, current glioma segmentation datasets focus on pre-treatment, diagnostic scans, where treatment effects and surgical cavities are not present. Thus, existing automatic segmentation models are not applicable to post-treatment imaging that is used for longitudinal evaluation of care. Here, we present a comparison of three-dimensional convolutional neural networks (nnU-Net architecture) trained on large temporally defined pre-treatment, post-treatment, and mixed cohorts. We used a total of 1563 imaging timepoints from 854 patients curated from 13 different institutions as well as diverse public data sets to understand the capabilities and limitations of automatic segmentation on glioma images with different phenotypic and treatment appearance. We assessed the performance of models using Dice coefficients on test cases from each group comparing predictions with manual segmentations generated by trained technicians. We demonstrate that training a combined model can be as effective as models trained on just one temporal group. The results highlight the importance of a diverse training set, that includes images from the course of disease and with effects from treatment, in the creation of a model that can accurately segment glioma MRIs at multiple treatment time points.

A Deep Convolutional Neural Network for Annotation of Magnetic Resonance Imaging Sequence Type.

The explosion of medical imaging data along with the advent of big data analytics has launched an exciting era for clinical research. One factor affecting the ability to aggregate large medical image collections for research is the lack of infrastructure for automated data annotation. Among all imaging modalities, annotation of magnetic resonance (MR) images is particularly challenging due to the non-standard labeling of MR image types. In this work, we aimed to train a deep neural network to annotate MR image sequence type for scans of brain tumor patients. We focused on the four most common MR sequence types within neuroimaging: T1-weighted (T1W), T1-weighted post-gadolinium contrast (T1Gd), T2-weighted (T2W), and T2-weighted fluid-attenuated inversion recovery (FLAIR). Our repository contains images acquired using a variety of pulse sequences, sequence parameters, field strengths, and scanner manufacturers. Image selection was agnostic to patient demographics, diagnosis, and the presence of tumor in the imaging field of view. We used a total of 14,400 two-dimensional images, each visualizing a different part of the brain. Data was split into train, validation, and test sets (9600, 2400, and 2400 images, respectively) and sets consisted of equal-sized groups of image types. Overall, the model reached an accuracy of 99% on the test set. Our results showed excellent performance of deep learning techniques in predicting sequence types for brain tumor MR images. We conclude deep learning models can serve as tools to support clinical research and facilitate efficient database management.

Integration of machine learning and mechanistic models accurately predicts variation in cell density of glioblastoma using multiparametric MRI.

Glioblastoma (GBM) is a heterogeneous and lethal brain cancer. These tumors are followed using magnetic resonance imaging (MRI), which is unable to precisely identify tumor cell invasion, impairing effective surgery and radiation planning. We present a novel hybrid model, based on multiparametric intensities, which combines machine learning (ML) with a mechanistic model of tumor growth to provide spatially resolved tumor cell density predictions. The ML component is an imaging data-driven graph-based semi-supervised learning model and we use the Proliferation-Invasion (PI) mechanistic tumor growth model. We thus refer to the hybrid model as the ML-PI model. The hybrid model was trained using 82 image-localized biopsies from 18 primary GBM patients with pre-operative MRI using a leave-one-patient-out cross validation framework. A Relief algorithm was developed to quantify relative contributions from the data sources. The ML-PI model statistically significantly outperformed (p < 0.001) both individual models, ML and PI, achieving a mean absolute predicted error (MAPE) of 0.106 ± 0.125 versus 0.199 ± 0.186 (ML) and 0.227 ± 0.215 (PI), respectively. Associated Pearson correlation coefficients for ML-PI, ML, and PI were 0.838, 0.518, and 0.437, respectively. The Relief algorithm showed the PI model had the greatest contribution to the result, emphasizing the importance of the hybrid model in achieving the high accuracy.

Brain MR Radiomics to Differentiate Cognitive Disorders.

OBJECTIVE: Subtle and gradual changes occur in the brain years before cognitive impairment due to age-related neurodegenerative disorders. The authors examined the utility of hippocampal texture analysis and volumetric features extracted from brain magnetic resonance (MR) data to differentiate between three cognitive groups (cognitively normal individuals, individuals with mild cognitive impairment, and individuals with Alzheimer's disease) and neuropsychological scores on the Clinical Dementia Rating (CDR) scale. METHODS: Data from 173 unique patients with 3-T T1-weighted MR images from the Alzheimer's Disease Neuroimaging Initiative database were analyzed. A variety of texture and volumetric features were extracted from bilateral hippocampal regions and were used to perform binary classification of cognitive groups and CDR scores. The authors used diagonal quadratic discriminant analysis in a leave-one-out cross-validation scheme. Sensitivity, specificity, and area under the receiver operating characteristic curve were used to assess the performance of models. RESULTS: The results show promise for hippocampal texture analysis to distinguish between no impairment and early stages of impairment. Volumetric features were more successful at differentiating between no impairment and advanced stages of impairment. CONCLUSIONS: MR radiomics may be a promising tool to classify various cognitive groups.

Canadian Association of Radiologists White Paper on Artificial Intelligence in Radiology.

Artificial intelligence (AI) is rapidly moving from an experimental phase to an implementation phase in many fields, including medicine. The combination of improved availability of large datasets, increasing computing power, and advances in learning algorithms has created major performance breakthroughs in the development of AI applications. In the last 5 years, AI techniques known as deep learning have delivered rapidly improving performance in image recognition, caption generation, and speech recognition. Radiology, in particular, is a prime candidate for early adoption of these techniques. It is anticipated that the implementation of AI in radiology over the next decade will significantly improve the quality, value, and depth of radiology's contribution to patient care and population health, and will revolutionize radiologists' workflows. The Canadian Association of Radiologists (CAR) is the national voice of radiology committed to promoting the highest standards in patient-centered imaging, lifelong learning, and research. The CAR has created an AI working group with the mandate to discuss and deliberate on practice, policy, and patient care issues related to the introduction and implementation of AI in imaging. This white paper provides recommendations for the CAR derived from deliberations between members of the AI working group. This white paper on AI in radiology will inform CAR members and policymakers on key terminology, educational needs of members, research and development, partnerships, potential clinical applications, implementation, structure and governance, role of radiologists, and potential impact of AI on radiology in Canada.

Computer-aided diagnosis of contrast-enhanced spectral mammography: A feasibility study.

OBJECTIVE: To evaluate whether the use of a computer-aided diagnosis-contrast-enhanced spectral mammography (CAD-CESM) tool can further increase the diagnostic performance of CESM compared with that of experienced radiologists. MATERIALS AND METHODS: This IRB-approved retrospective study analyzed 50 lesions described on CESM from August 2014 to December 2015. Histopathologic analyses, used as the criterion standard, revealed 24 benign and 26 malignant lesions. An expert breast radiologist manually outlined lesion boundaries on the different views. A set of morphologic and textural features were then extracted from the low-energy and recombined images. Machine-learning algorithms with feature selection were used along with statistical analysis to reduce, select, and combine features. Selected features were then used to construct a predictive model using a support vector machine (SVM) classification method in a leave-one-out-cross-validation approach. The classification performance was compared against the diagnostic predictions of 2 breast radiologists with access to the same CESM cases. RESULTS: Based on the SVM classification, CAD-CESM correctly identified 45 of 50 lesions in the cohort, resulting in an overall accuracy of 90%. The detection rate for the malignant group was 88% (3 false-negative cases) and 92% for the benign group (2 false-positive cases). Compared with the model, radiologist 1 had an overall accuracy of 78% and a detection rate of 92% (2 false-negative cases) for the malignant group and 62% (10 false-positive cases) for the benign group. Radiologist 2 had an overall accuracy of 86% and a detection rate of 100% for the malignant group and 71% (8 false-positive cases) for the benign group. CONCLUSIONS: The results of our feasibility study suggest that a CAD-CESM tool can provide complementary information to radiologists, mainly by reducing the number of false-positive findings.

Computed Tomography-Based Texture Analysis to Determine Human Papillomavirus Status of Oropharyngeal Squamous Cell Carcinoma.

OBJECTIVE: To determine whether machine learning can accurately classify human papillomavirus (HPV) status of oropharyngeal squamous cell carcinoma (OPSCC) using computed tomography (CT)-based texture analysis. METHODS: Texture analyses were retrospectively applied to regions of interest from OPSCC primary tumors on contrast-enhanced neck CT, and machine learning was used to create a model that classified HPV status with the highest accuracy. Results were compared against the blinded review of 2 neuroradiologists. RESULTS: The HPV-positive (n = 92) and -negative (n = 15) cohorts were well matched clinically. Neuroradiologist classification accuracies for HPV status (44.9%, 55.1%) were not significantly different (P = 0.13), and there was a lack of agreement between the 2 neuroradiologists (κ = -0.145). The best machine learning model had an accuracy of 75.7%, which was greater than either neuroradiologist (P < 0.001, P = 0.002). CONCLUSIONS: Useful diagnostic information regarding HPV infection can be extracted from the CT appearance of OPSCC beyond what is apparent to the trained human eye.

Radiogenomics to characterize regional genetic heterogeneity in glioblastoma.

BACKGROUND: Glioblastoma (GBM) exhibits profound intratumoral genetic heterogeneity. Each tumor comprises multiple genetically distinct clonal populations with different therapeutic sensitivities. This has implications for targeted therapy and genetically informed paradigms. Contrast-enhanced (CE)-MRI and conventional sampling techniques have failed to resolve this heterogeneity, particularly for nonenhancing tumor populations. This study explores the feasibility of using multiparametric MRI and texture analysis to characterize regional genetic heterogeneity throughout MRI-enhancing and nonenhancing tumor segments. METHODS: We collected multiple image-guided biopsies from primary GBM patients throughout regions of enhancement (ENH) and nonenhancing parenchyma (so called brain-around-tumor, [BAT]). For each biopsy, we analyzed DNA copy number variants for core GBM driver genes reported by The Cancer Genome Atlas. We co-registered biopsy locations with MRI and texture maps to correlate regional genetic status with spatially matched imaging measurements. We also built multivariate predictive decision-tree models for each GBM driver gene and validated accuracies using leave-one-out-cross-validation (LOOCV). RESULTS: We collected 48 biopsies (13 tumors) and identified significant imaging correlations (univariate analysis) for 6 driver genes: EGFR, PDGFRA, PTEN, CDKN2A, RB1, and TP53. Predictive model accuracies (on LOOCV) varied by driver gene of interest. Highest accuracies were observed for PDGFRA (77.1%), EGFR (75%), CDKN2A (87.5%), and RB1 (87.5%), while lowest accuracy was observed in TP53 (37.5%). Models for 4 driver genes (EGFR, RB1, CDKN2A, and PTEN) showed higher accuracy in BAT samples (n = 16) compared with those from ENH segments (n = 32). CONCLUSION: MRI and texture analysis can help characterize regional genetic heterogeneity, which offers potential diagnostic value under the paradigm of individualized oncology.

Adaptive noise correction of dual-energy computed tomography images.

PURPOSE: Noise reduction in material density images is a necessary preprocessing step for the correct interpretation of dual-energy computed tomography (DECT) images. In this paper we describe a new method based on a local adaptive processing to reduce noise in DECT images METHODS: An adaptive neighborhood Wiener (ANW) filter was implemented and customized to use local characteristics of material density images. The ANW filter employs a three-level wavelet approach, combined with the application of an anisotropic diffusion filter. Material density images and virtual monochromatic images are noise corrected with two resulting noise maps. RESULTS: The algorithm was applied and quantitatively evaluated in a set of 36 images. From that set of images, three are shown here, and nine more are shown in the online supplementary material. Processed images had higher signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) than the raw material density images. The average improvements in SNR and CNR for the material density images were 56.5 and 54.75%, respectively. CONCLUSION: We developed a new DECT noise reduction algorithm. We demonstrate throughout a series of quantitative analyses that the algorithm improves the quality of material density images and virtual monochromatic images.

Multi-Parametric MRI and Texture Analysis to Visualize Spatial Histologic Heterogeneity and Tumor Extent in Glioblastoma.

BACKGROUND: Genetic profiling represents the future of neuro-oncology but suffers from inadequate biopsies in heterogeneous tumors like Glioblastoma (GBM). Contrast-enhanced MRI (CE-MRI) targets enhancing core (ENH) but yields adequate tumor in only ~60% of cases. Further, CE-MRI poorly localizes infiltrative tumor within surrounding non-enhancing parenchyma, or brain-around-tumor (BAT), despite the importance of characterizing this tumor segment, which universally recurs. In this study, we use multiple texture analysis and machine learning (ML) algorithms to analyze multi-parametric MRI, and produce new images indicating tumor-rich targets in GBM. METHODS: We recruited primary GBM patients undergoing image-guided biopsies and acquired pre-operative MRI: CE-MRI, Dynamic-Susceptibility-weighted-Contrast-enhanced-MRI, and Diffusion Tensor Imaging. Following image coregistration and region of interest placement at biopsy locations, we compared MRI metrics and regional texture with histologic diagnoses of high- vs low-tumor content (≥80% vs <80% tumor nuclei) for corresponding samples. In a training set, we used three texture analysis algorithms and three ML methods to identify MRI-texture features that optimized model accuracy to distinguish tumor content. We confirmed model accuracy in a separate validation set. RESULTS: We collected 82 biopsies from 18 GBMs throughout ENH and BAT. The MRI-based model achieved 85% cross-validated accuracy to diagnose high- vs low-tumor in the training set (60 biopsies, 11 patients). The model achieved 81.8% accuracy in the validation set (22 biopsies, 7 patients). CONCLUSION: Multi-parametric MRI and texture analysis can help characterize and visualize GBM's spatial histologic heterogeneity to identify regional tumor-rich biopsy targets.

Pilot study: Evaluation of dual-energy computed tomography measurement strategies for positron emission tomography correlation in pancreatic adenocarcinoma.

We sought to determine whether dual-energy computed tomography (DECT) measurements correlate with positron emission tomography (PET) standardized uptake values (SUVs) in pancreatic adenocarcinoma, and to determine the optimal DECT imaging variables and modeling strategy to produce the highest correlation with maximum SUV (SUVmax). We reviewed 25 patients with unresectable pancreatic adenocarcinoma seen at Mayo Clinic, Scottsdale, Arizona, who had PET-computed tomography (PET/CT) and enhanced DECT performed the same week between March 25, 2010 and December 9, 2011. For each examination, DECT measurements were taken using one of three methods: (1) average values of three tumor regions of interest (ROIs) (method 1); (2) one ROI in the area of highest subjective DECT enhancement (method 2); and (3) one ROI in the area corresponding to PET SUVmax (method 3). There were 133 DECT variables using method 1, and 89 using the other methods. Univariate and multivariate analysis regression models were used to identify important correlations between DECT variables and PET SUVmax. Both R2 and adjusted R2 were calculated for the multivariate model to compensate for the increased number of predictors. The average SUVmax was 5 (range, 1.8-12.0). Multivariate analysis of DECT imaging variables outperformed univariate analysis (r = 0.91; R2 = 0.82; adjusted R2 = 0.75 vs. r < 0.58; adjusted R2 < 0.34). Method 3 had the highest correlation with PET SUVmax (R2 = 0.82), followed by method 1 (R2 = 0.79) and method 2 R2 = 0.57). DECT thus has clinical potential as a surrogate for, or as a complement to, PET in patients with pancreatic adenocarcinoma.

Validity of the Apple iPhone® /iPod Touch® as an accelerometer-based physical activity monitor: a proof-of-concept study.

BACKGROUND: The popularity of smartphones has led researchers to ask if they can replace traditional tools for assessing free-living physical activity. Our purpose was to establish proof-of-concept that a smartphone could record acceleration during physical activity, and those data could be modeled to predict activity type (walking or running), speed (km·h-1), and energy expenditure (METs). METHODS: An application to record and e-mail accelerations was developed for the Apple iPhone®/iPod Touch®. Twenty-five healthy adults performed treadmill walking (4.0 km·h-1 to 7.2 km·h-1) and running (8.1 km·h-1 to 11.3 km·h-1) wearing the device. Criterion energy expenditure measurements were collected via metabolic cart. RESULTS: Activity type was classified with 99% accuracy. Speed was predicted with a bias of 0.02 km·h-1 (SEE: 0.57 km·h-1) for walking, -0.03 km·h-1 (SEE: 1.02 km·h-1) for running. Energy expenditure was predicted with a bias of 0.35 METs (SEE: 0.75 METs) for walking, -0.43 METs (SEE: 1.24 METs) for running. CONCLUSION: Our results suggest that an iPhone/iPod Touch can predict aspects of locomotion with accuracy similar to other accelerometer-based tools. Future studies may leverage this and the additional features of smartphones to improve data collection and compliance.

Active inflammation increases the heterogeneity of MRI texture in mice with relapsing experimental allergic encephalomyelitis.

Inflammation modulates tissue damage in relapsing-remitting multiple sclerosis (MS) both acutely and chronically, but its severity is difficult to evaluate with conventional MRI analysis. In mice with experimental allergic encephalomyelitis (EAE, a model of MS), we administered ultra small particles of iron oxide to track macrophage-mediated inflammation during the onset (relapse) and recovery (remission) of disease activity using high field MRI. We performed MRI texture analysis, a sensitive measure of tissue regularity, and T2 assessment both in EAE lesions and the control tissue, and measured spinal cord volume. We found that inflammation was 3 times more remarkable at onset than at recovery of EAE in histology yet demyelination appeared similar across animals and disease course. In MRI, lesion texture was more heterogeneous; T2 was lower; and spinal cord volume was greater in EAE than in controls, but only MRI texture was worse at relapse than at remission of EAE. Moreover, MRI texture correlated with spinal cord volume and tended to correlate with the extent of disability in EAE. While subject to further confirmation, our findings may suggest the sensitivity of MRI texture analysis for accessing inflammation.

Validation study of a fast, accurate, and precise brain tumor volume measurement.

UNLABELLED: Precision and accuracy are sometimes sacrificed to ensure that medical image processing is rapid. To address this, our lab had developed a novel level set segmentation algorithm that is 16× faster and >96% accurate on realistic brain phantoms. METHODS: This study reports speed, precision and estimated accuracy of our algorithm when measuring MRIs of meningioma brain tumors and compares it to manual tracing and modified MacDonald (MM) ellipsoid criteria. A repeated-measures study allowed us to determine measurement precisions (MPs) - clinically relevant thresholds for statistically significant change. RESULTS: Speed: the level set, MM, and trace methods required 1:20, 1:35, and 9:35 (mm:ss) respectively on average to complete a volume measurement (p<0.05). Accuracy: the level set was not statistically different to the estimated true lesion volumes (p>0.05). Precision: the MM's within-operator and between-operator MPs were significantly higher (worse) than the other methods (p<0.05). The observed difference in MP between the level set and trace methods did not reach statistical significance (p>0.05). CONCLUSION: Our level set is faster on average than MM, yet has accuracy and precision comparable to manual tracing.

A smartphone client-server teleradiology system for primary diagnosis of acute stroke.

BACKGROUND: Recent advances in the treatment of acute ischemic stroke have made rapid acquisition, visualization, and interpretation of images a key factor for positive patient outcomes. We have developed a new teleradiology system based on a client-server architecture that enables rapid access to interactive advanced 2-D and 3-D visualization on a current generation smartphone device (Apple iPhone or iPod Touch, or an Android phone) without requiring patient image data to be stored on the device. Instead, a server loads and renders the patient images, then transmits a rendered frame to the remote device. OBJECTIVE: Our objective was to determine if a new smartphone client-server teleradiology system is capable of providing accuracies and interpretation times sufficient for diagnosis of acute stroke. METHODS: This was a retrospective study. We obtained 120 recent consecutive noncontrast computed tomography (NCCT) brain scans and 70 computed tomography angiogram (CTA) head scans from the Calgary Stroke Program database. Scans were read by two neuroradiologists, one on a medical diagnostic workstation and an iPod or iPhone (hereafter referred to as an iOS device) and the other only on an iOS device. NCCT brain scans were evaluated for early signs of infarction, which includes early parenchymal ischemic changes and dense vessel sign, and to exclude acute intraparenchymal hemorrhage and stroke mimics. CTA brain scans were evaluated for any intracranial vessel occlusion. The interpretations made on an iOS device were compared with those made at a workstation. The total interpretation times were recorded for both platforms. Interrater agreement was assessed. True positives, true negatives, false positives, and false negatives were obtained, and sensitivity, specificity, and accuracy of detecting the abnormalities on the iOS device were computed. RESULTS: The sensitivity, specificity, and accuracy of detecting intraparenchymal hemorrhage were 100% using the iOS device with a perfect interrater agreement (kappa=1). The sensitivity, specificity, and accuracy of detecting acute parenchymal ischemic change were 94.1%, 100%, and 98.09% respectively for reader 1 and 97.05%, 100%, and 99.04% for reader 2 with nearly perfect interrater agreement (kappa=.8). The sensitivity, specificity, and accuracy of detecting dense vessel sign were 100%, 95.4%, and 96.19% respectively for reader 1 and 72.2%, 100%, and 95.23% for reader 2 using the iOS device with a good interrater agreement (kappa=.69). The sensitivity, specificity, and accuracy of detecting vessel occlusion on CT angiography scans were 94.4%, 100%, and 98.46% respectively for both readers using the iOS device, with perfect interrater agreement (kappa=1). No significant difference (P<.05) was noted in the interpretation time between the workstation and iOS device. CONCLUSION: The smartphone client-server teleradiology system appears promising and may have the potential to allow urgent management decisions in acute stroke. However, this study was retrospective, involved relatively few patient studies, and only two readers. Generalizing conclusions about its clinical utility, especially in other diagnostic use cases, should not be made until additional studies are performed.

iPhone-based teleradiology for the diagnosis of acute cervico-dorsal spine trauma.

OBJECTIVE: To assess the feasibility of iPhone-based teleradiology as a potential solution for the diagnosis of acute cervico-dorsal spine trauma. MATERIALS AND METHODS: We have developed a solution that allows visualization of images on the iPhone. Our system allows rapid, remote, secure, visualization of medical images without storing patient data on the iPhone. This retrospective study is comprised of cervico-dorsal computed tomogram (CT) scan examination of 75 consecutive patients having clinically suspected cervico-dorsal spine fracture. Two radiologists reviewed CT scan images on the iPhone. Computed tomogram spine scans were analyzed for vertebral body fracture and posterior elements fractures, any associated subluxation-dislocation and cord lesion. The total time taken from the launch of viewing application on the iPhone until interpretation was recorded. The results were compared with that of a diagnostic workstation monitor. Inter-rater agreement was assessed. RESULTS: The sensitivity and accuracy of detecting vertebral body fractures was 80% and 97% by both readers using the iPhone system with a perfect inter-rater agreement (kappa:1). The sensitivity and accuracy of detecting posterior elements fracture was 75% and 98% for Reader 1 and 50% and 97% for Reader 2 using the iPhone. There was good inter-rater agreement (kappa: 0.66) between both readers. No statistically significant difference was noted between time on the workstation and the iPhone system. CONCLUSION: iPhone-based teleradiology system is accurate in the diagnosis of acute cervicodorsal spinal trauma. It allows rapid, remote, secure, visualization of medical images without storing patient data on the iPhone.

AnalyzeNNLS: magnetic resonance multiexponential decay image analysis.

Exponential decays are fundamental to magnetic resonance imaging, yet adequately sampling and analyzing multiexponential decays is rarely attempted. The advantage of multiexponential analysis is the quantification of sub-voxel structure caused by water compartmentalization, with application as a non-invasive imaging biomarker for myelin. We have developed AnalyzeNNLS, software designed specifically for multiexponential decay image analysis that has a user-friendly graphical user interface and can analyze data from many MR manufacturers. AnalyzeNNLS is a simple, platform independent analysis tool that was created using the extensive mathematical and visualization libraries in Matlab, and released as open source code allowing scientists to evaluate, scrutinize, improve, and expand.

An analysis of image texture, tumor location, and MGMT promoter methylation in glioblastoma using magnetic resonance imaging.

In glioblastoma (GBM), promoter methylation of the DNA repair gene O(6)-methylguanine-DNA methyltransferase (MGMT) is associated with benefit from chemotherapy. Correlations between MGMT promoter methylation and visually assessed imaging features on magnetic resonance (MR) have been reported suggesting that noninvasive detection of MGMT methylation status might be possible. Our study assessed whether MGMT methylation status in GBM could be predicted using MR imaging. We conducted a retrospective analysis of MR images in patients with newly diagnosed GBM. Tumor texture was assessed by two methods. First, we analyzed texture by expert consensus describing the tumor borders, presence or absence of cysts, pattern of enhancement, and appearance of tumor signal in T2-weighted images. Then, we applied space-frequency texture analysis based on the S-transform. Tumor location within the brain was determined using automatized image registration and segmentation techniques. Their association with MGMT methylation was analyzed. We confirmed that ring enhancement assessed visually is significantly associated with unmethylated MGMT promoter status (P=0.006). Texture features on T2-weighted images assessed by the space-frequency analysis were significantly different between methylated and unmethylated cases (P<0.05). However, blinded classification of MGMT promoter methylation status reached an accuracy of only 71%. There were no significant differences in the locations of methylated and unmethylated GBM tumors. Our results provide further evidence that individual MR features are associated with MGMT methylation but better algorithms for predicting methylation status are needed. The relevance of this study lies on the application of novel techniques for the analysis of anatomical MR images of patients with GBM allowing the evaluation of subtleties not seen by an observer and facilitating the standardization of the methods, decreasing the potential for interobserver bias.

Quantitative T2 analysis: the effects of noise, regularization, and multivoxel approaches.

Typical quantitative T2 (qT2) analysis involves creating T2 distributions using a regularized algorithm from region-of-interest averaged decay data. This study uses qT2 analysis of simulated and experimental decay signals to determine how (a) noise-type, (b) regularization, and (c) region-of-interest versus multivoxel analyses affect T2 distributions. Our simulations indicate that regularization causes myelin water fraction and intra/extracellular water geometric mean T2 underestimation that worsens as the signal-to-noise ratio decreases. The underestimation was greater for intra/extracellular water geometric mean T2 measures using Rician noise. Simulations showed significant differences between myelin water fractions determined using region-of-interest and multivoxel approaches compared to the true value. The nonregularized voxel-based approach gave the most accurate measure of myelin water fraction and intra/extracellular water geometric mean T2 for a given signal-to-noise ratio and noise type. Additionally, multivoxel analysis provides important information about the variability of the analysis. Results obtained from in vivo rat data were similar to our simulation results. In each case, a nonregularized, multivoxel analysis provided myelin water fractions significantly different from the regularized approaches and obtained the largest myelin water fraction. We conclude that quantitative T2 analysis is best performed using a nonregularized, multivoxel approach.

Multiexponential T2 and magnetization transfer MRI of demyelination and remyelination in murine spinal cord.

Identification of remyelination is important in the evaluation of potential treatments of demyelinating diseases such as multiple sclerosis. Local injection of lysolecithin into the brain or spinal cord provides a murine model of demyelination with spontaneous remyelination. The aim of this study was to determine if quantitative, multicomponent T(2) (qT(2)) analysis and magnetization transfer ratio (MTR), both indicative of myelin content, could detect changes in myelination, particularly remyelination, of the cervical spinal cord in mice treated with lysolecithin. We found that the myelin water fraction and geometric mean T(2) value of the intra/extracellular water significantly decreased at 14 days then returned to control levels by 28 days after injury, corresponding to clearance of myelin debris and remyelination which was shown by eriochrome cyanine and oil red O staining of histological sections. The MTR was significantly decreased 14 days after lysolecithin injection, and remained low over the time course studied. Evidence of demyelination shown by both qT(2) and MTR lagged behind the histological evidence of demyelination. Myelin water fraction increased with remyelination, however MTR remained lower after 28 days. The difference between qT(2) and MTR may identify early remyelination.