Multiparametric analysis of magnetic resonance images for glioma grading and patient survival time prediction

Magnetic Resonance Imaging of Primary Adult Brain Tumors: State of the Art and Future Perspectives

MRI is undoubtedly the cornerstone of brain tumor imaging, playing a key role in all phases of patient management, starting from diagnosis, through therapy planning, to treatment response and/or recurrence assessment. Currently, neuroimaging can describe morphologic and non-morphologic (functional, hemodynamic, metabolic, cellular, microstructural, and sometimes even genetic) characteristics of brain tumors, greatly contributing to diagnosis and follow-up. Knowing the technical aspects, strength and limits of each MR technique is crucial to correctly interpret MR brain studies and to address clinicians to the best treatment strategy. This article aimed to provide an overview of neuroimaging in the assessment of adult primary brain tumors. We started from the basilar role of conventional/morphological MR sequences, then analyzed, one by one, the non-morphological techniques, and finally highlighted future perspectives, such as radiomics and artificial intelligence.

A machine learning-based survival prediction model of high grade glioma by integration of clinical and dose-volume histogram parameters

PURPOSE

Glioma is the most common type of primary brain tumor in adults, and it causes significant morbidity and mortality, especially in high-grade glioma (HGG) patients. The accurate prognostic prediction of HGG is vital and helpful for clinicians when developing therapeutic strategies. Therefore, we propose a machine learning-based survival prediction model by analyzing clinical and dose-volume histogram (DVH) parameters, to improve the performance of the risk model in HGG patients.

METHODS

Eight clinical variables and 39 DVH parameters were extracted for each patient, who received radiotherapy for HGG with active follow-up. Ninety-five patients were randomly divided into training and testing cohorts, and we employed random survival forest (RSF), support vector machine (SVM), and Cox proportional hazards (CPHs) models to predict survival. Calibration plots, concordance indexes, and decision curve analyses were used to evaluate the calibration, discrimination, and clinical utility of these three models.

RESULTS

The RSF model showed the best performance among the three models, with concordance indexes of 0.824 and 0.847 in the training and testing sets, respectively, followed by the SVM (0.792/0.823) and CPH (0.821/0.811) models. Specifically, in the RSF model, we identified age, gross tumor volume (GTV), grade, Karnofsky performance status (KPS), isocitrate dehydrogenase (IDH), and D99 as important variables associated with survival. The AUCs of the testing set were 92.4%, 87.7%, and 84.0% for 1-, 2-, and 3-year survival, respectively. According to this model, HGG patients can be divided into high- and low-risk groups.

CONCLUSION

The machine learning-based RSF model integrating both clinical and DVH variables is an improved and useful tool for predicting the survival of HGG patients.

Extracellular Vesicles from Neurosurgical Aspirates Identifies Chaperonin Containing TCP1 Subunit 6A as a Potential Glioblastoma Biomarker with Prognostic Significance

Glioblastoma, WHO-grade IV glioma, carries a dismal prognosis owing to its infiltrative growth and limited treatment options. Glioblastoma-derived extracellular vesicles (EVs; 30-1000 nm membranous particles) influence the microenvironment to mediate tumor aggressiveness and carry oncogenic cargo across the blood-brain barrier into the circulation. As such, EVs are biomarker reservoirs with enormous potential for assessing glioblastoma tumors in situ. Neurosurgical aspirates are rich sources of EVs, isolated directly from glioma microenvironments. EV proteomes enriched from glioblastoma (n = 15) and glioma grade II-III (n = 7) aspirates are compared and 298 differentially-abundant proteins (p-value < 0.00496) are identified using quantitative LC-MS/MS. Along with previously reported glioblastoma-associated biomarkers, levels of all eight subunits of the key molecular chaperone, T-complex protein 1 Ring complex (TRiC), are higher in glioblastoma-EVs, including CCT2, CCT3, CCT5, CCT6A, CCT7, and TCP1 (p < 0.00496). Analogous increases in TRiC transcript levels and DNA copy numbers are detected in silico; CCT6A has the greatest induction of expression and amplification in glioblastoma and shows a negative association with survival (p = 0.006). CCT6A is co-localized with EGFR at 7p11.2, with a strong tendency for co-amplification (p < 0.001). Immunohistochemistry corroborates the CCT6A proteomics measurements and indicated a potential link between EGFR and CCT6A tissue expression. Putative EV-biomarkers described here should be further assessed in peripheral blood.

Prediction of malignant glioma grades using contrast-enhanced T1-weighted and T2-weighted magnetic resonance images based on a radiomic analysis

We conducted a feasibility study to predict malignant glioma grades via radiomic analysis using contrast-enhanced T1-weighted magnetic resonance images (CE-T1WIs) and T2-weighted magnetic resonance images (T2WIs). We proposed a framework and applied it to CE-T1WIs and T2WIs (with tumor region data) acquired preoperatively from 157 patients with malignant glioma (grade III: 55, grade IV: 102) as the primary dataset and 67 patients with malignant glioma (grade III: 22, grade IV: 45) as the validation dataset. Radiomic features such as size/shape, intensity, histogram, and texture features were extracted from the tumor regions on the CE-T1WIs and T2WIs. The Wilcoxon-Mann-Whitney (WMW) test and least absolute shrinkage and selection operator logistic regression (LASSO-LR) were employed to select the radiomic features. Various machine learning (ML) algorithms were used to construct prediction models for the malignant glioma grades using the selected radiomic features. Leave-one-out cross-validation (LOOCV) was implemented to evaluate the performance of the prediction models in the primary dataset. The selected radiomic features for all folds in the LOOCV of the primary dataset were used to perform an independent validation. As evaluation indices, accuracies, sensitivities, specificities, and values for the area under receiver operating characteristic curve (or simply the area under the curve (AUC)) for all prediction models were calculated. The mean AUC value for all prediction models constructed by the ML algorithms in the LOOCV of the primary dataset was 0.902 ± 0.024 (95% CI (confidence interval), 0.873-0.932). In the independent validation, the mean AUC value for all prediction models was 0.747 ± 0.034 (95% CI, 0.705-0.790). The results of this study suggest that the malignant glioma grades could be sufficiently and easily predicted by preparing the CE-T1WIs, T2WIs, and tumor delineations for each patient. Our proposed framework may be an effective tool for preoperatively grading malignant gliomas.

Relevance of calcification and contrast enhancement pattern for molecular diagnosis and survival prediction of gliomas based on the 2016 World Health Organization Classification

OBJECTIVES

The significance of conventional neuroimaging features for predicting molecular diagnosis and patient survival based on the updated World Health Organization (WHO) classification remains uncertain. We assessed the relevance of neuroimaging features (ring enhancement [RE], non-ring enhancement [non-RE], overall gadolinium enhancement [GdE], and intratumoral calcification [IC]) for molecular diagnosis and survival in glioma patients.

PATIENTS AND METHODS

We evaluated 234 glioma patients according to the updated WHO classification. Isocitrate dehydrogenase (IDH), H3F3A, BRAF hotspot mutations, TERT promotor mutation, and chromosome 1p/19q co-deletion were examined. RE, non-RE, GdE, and IC were evaluated as significant neuroimaging findings. Kaplan-Meier analyses were performed to evaluate overall survival (OS) and the correlations of prognostic factors were evaluated by log-rank tests. Univariate and multivariate analyses were performed to detect prognostic factors for OS.

RESULTS

A total of 207 patients were eligible. In 110 patients presenting RE, 102 (93%) were glioblastoma (GBM), IDH-wild type. In 97 patients without RE, presence of GdE or IC were not significantly different between IDH-mutant and -wild type tumors, whereas presence of GdE was a significant indicator of higher WHO grades. IC was the only significant finding for 1p/19q co-deleted tumors. TERT promoter mutation was observed in 7/17 patients with diffuse astrocytic glioma, IDH-wild type; recently-defined as "molecular GBM." IC, RE, and GdE were observed with lower prevalence in molecular GBMs. While presence of RE, GdE, and absence of IC were significant factors of OS in overall cohort, presence of GdE was not significant in OS in cases without RE, and IDH-mutant tumors. IC was a significant predictor of favorable OS in cases without RE and IDH-wild type tumors. Multivariate analysis also validated these findings.

CONCLUSION

GdE alone is not a significant predictor of IDH mutation status, but the pattern of enhancement is a significant predictor with RE demonstrating high sensitivity and specificity for GBM, IDH-wild type. Predicting "molecular GBM" by conventional neuroimaging is difficult. Moreover, GdE is not a significant factor of survival analyzed with pattern of enhancement or molecular stratifications. IC is an important radiographic finding for predicting molecular diagnosis and survival in glioma patients.

Multimodal 3D DenseNet for IDH Genotype Prediction in Gliomas

Non-invasive prediction of isocitrate dehydrogenase (IDH) genotype plays an important role in tumor glioma diagnosis and prognosis. Recently, research has shown that radiology images can be a potential tool for genotype prediction, and fusion of multi-modality data by deep learning methods can further provide complementary information to enhance prediction accuracy. However, it still does not have an effective deep learning architecture to predict IDH genotype with three-dimensional (3D) multimodal medical images. In this paper, we proposed a novel multimodal 3D DenseNet (M3D-DenseNet) model to predict IDH genotypes with multimodal magnetic resonance imaging (MRI) data. To evaluate its performance, we conducted experiments on the BRATS-2017 and The Cancer Genome Atlas breast invasive carcinoma (TCGA-BRCA) dataset to get image data as input and gene mutation information as the target, respectively. We achieved 84.6% accuracy (area under the curve (AUC) = 85.7%) on the validation dataset. To evaluate its generalizability, we applied transfer learning techniques to predict World Health Organization (WHO) grade status, which also achieved a high accuracy of 91.4% (AUC = 94.8%) on validation dataset. With the properties of automatic feature extraction, and effective and high generalizability, M3D-DenseNet can serve as a useful method for other multimodal radiogenomics problems and has the potential to be applied in clinical decision making.

Contrast enhancement predicting survival in integrated molecular subtypes of diffuse glioma: an observational cohort study

INTRODUCTION

To assess the predictive value of magnetic resonance imaging (MRI) gadolinium enhancement as a prognostic factor in the 2016 World Health Organization Classification of Tumors of the Central Nervous System integrated glioma groups.

METHODS

Four-hundred fifty patients with histopathologically confirmed glioma were retrospectively assessed between 07/1997 and 06/2014 using gadolinium enhancement, survival, and relevant prognostic molecular data [isocitrate dehydrogenase (IDH); alpha-thalassemia/mental retardation syndrome X-linked (ATRX); chromosome 1p/19q loss of heterozygosity; and O6-methylguanine DNA methyltransferase (MGMT)]. The Kaplan-Meier method was used to assess univariate survival data. A multivariate Cox proportional hazards model was performed on significant results from the univariate analysis.

RESULTS

There were significant differences in survival between patient age (p < 0.0001), WHO glioma grades (p < 0.0001), and integrated molecular profiles (p < 0.0001). Patients with IDH1/2 mutation, loss of ATRX expression, and methylated MGMT promoter showed significantly better survival than those with the IDH_wild-type (p < 0.0001), retained ATRX expression (p < 0.0001), and unmethylated MGMT promoter (p = 0.019). Survival was significantly better in patients without gadolinium enhancement (p = 0.009) who were in the IDH_wild-type glioma and glioma with retained ATRX expression groups (p = 0.018 and 0.030, respectively).

CONCLUSIONS

In univariate analysis, the presence of gadolinium enhancement on preoperative MRI scans is an unfavorable factor for survival. Regarding the molecular subgroups, gadolinium enhancement is an unfavorable prognostic factor in gliomas with IDH_wild-type and those with ATRX retention. However, in multivariate analysis only patient age, IDH1/2 mutation status, MGMT promoter methylation status, and WHO grade IV are relevant for predicting survival.

Promoting Collaborations Between Radiologists and Scientists

Radiology as a discipline thrives on the dynamic interplay between technological and clinical advances. Progress in almost all facets of the imaging sciences is highly dependent on complex tools sourced from physics, engineering, biology, and the clinical sciences to obtain, process, and view imaging studies. The application of these tools, however, requires broad and deep medical knowledge about disease pathophysiology and its relationship with medical imaging. This relationship between clinical medicine and imaging technology, nurtured and fostered over the past 75 years, has cultivated extraordinarily rich collaborative opportunities between basic scientists, engineers, and physicians. In this review, we attempt to provide a framework to identify both currently successful collaborative ventures and future opportunities for scientific partnership. This invited review is a product of a special working group within the Association of University Radiologists-Radiology Research Alliance.

Predictive modeling of outcomes following definitive chemoradiotherapy for oropharyngeal cancer based on FDG-PET image characteristics

In this study, we investigate the use of imaging feature-based outcomes research ('radiomics') combined with machine learning techniques to develop robust predictive models for the risk of all-cause mortality (ACM), local failure (LF), and distant metastasis (DM) following definitive chemoradiation therapy (CRT). One hundred seventy four patients with stage III-IV oropharyngeal cancer (OC) treated at our institution with CRT with retrievable pre- and post-treatment 18F-fluorodeoxyglucose positron emission tomography (FDG-PET) scans were identified. From pre-treatment PET scans, 24 representative imaging features of FDG-avid disease regions were extracted. Using machine learning-based feature selection methods, multiparameter logistic regression models were built incorporating clinical factors and imaging features. All model building methods were tested by cross validation to avoid overfitting, and final outcome models were validated on an independent dataset from a collaborating institution. Multiparameter models were statistically significant on 5 fold cross validation with the area under the receiver operating characteristic curve (AUC)  =  0.65 (p  =  0.004), 0.73 (p  =  0.026), and 0.66 (p  =  0.015) for ACM, LF, and DM, respectively. The model for LF retained significance on the independent validation cohort with AUC  =  0.68 (p  =  0.029) whereas the models for ACM and DM did not reach statistical significance, but resulted in comparable predictive power to the 5 fold cross validation with AUC  =  0.60 (p  =  0.092) and 0.65 (p  =  0.062), respectively. In the largest study of its kind to date, predictive features including increasing metabolic tumor volume, increasing image heterogeneity, and increasing tumor surface irregularity significantly correlated to mortality, LF, and DM on 5 fold cross validation in a relatively uniform single-institution cohort. The LF model also retained significance in an independent population.

DCE-MRI, DW-MRI, and MRS in Cancer: Challenges and Advantages of Implementing Qualitative and Quantitative Multi-parametric Imaging in the Clinic

Multi-parametric magnetic resonance imaging (mpMRI) offers a unique insight into tumor biology by combining functional MRI techniques that inform on cellularity (diffusion-weighted MRI), vascular properties (dynamic contrast-enhanced MRI), and metabolites (magnetic resonance spectroscopy) and has scope to provide valuable information for prognostication and response assessment. Challenges in the application of mpMRI in the clinic include the technical considerations in acquiring good quality functional MRI data, development of robust techniques for analysis, and clinical interpretation of the results. This article summarizes the technical challenges in acquisition and analysis of multi-parametric MRI data before reviewing the key applications of multi-parametric MRI in clinical research and practice.

Quantitative Clinical Imaging Methods for Monitoring Intratumoral Evolution

Solid tumors are multiscale, open, complex, dynamic systems: complex because they have many interacting components, dynamic because both the components and their interactions can change with time, and open because the tumor freely communicates with surrounding and even distant host tissue. Thus, it is not surprising that striking intratumoral variations are commonly observed in clinical imaging such as MRI and CT and that several recent studies found striking regional variations in the molecular properties of cancer cells from the same tumor. Interestingly, this spatial heterogeneity in molecular properties of tumor cells is typically ascribed to branching clonal evolution due to accumulating mutations while macroscopic variations observed in, for example, clinical MRI scans are usually viewed as functions of blood flow. The clinical significance of spatial heterogeneity has not been fully determined but there is a general consensus that the varying intratumoral landscape along with patient factors such as age, morbidity and lifestyle, contributes significantly to the often unpredictable response of individual patients within a disease cohort treated with the same standard-of-care therapy.Here we investigate the potential link between macroscopic tumor heterogeneity observed by clinical imaging and spatial variations in the observed molecular properties of cancer cells. We build on techniques developed in landscape ecology to link regional variations in the distribution of species with local environmental conditions that define their habitat. That is, we view each region of the tumor as a local ecosystem consisting of environmental conditions such as access to nutrients, oxygen, and means of waste clearance related to blood flow and the local population of tumor cells that both adapt to these conditions and, to some extent, change them through, for example, production of angiogenic factors. Furthermore, interactions among neighboring habitats can produce broader regional dynamics so that the internal diversity of tumors is the net result of complex multiscale somatic Darwinian interactions.Methods in landscape ecology harness Darwinian dynamics to link the environmental properties of a given region to the local populations which are assumed to represent maximally fit phenotypes within those conditions. Consider a common task of a landscape ecologist: defining the spatial distribution of species in a large region, e.g., in a satellite image. Clearly the most accurate approach requires a meter by meter survey of the multiple square kilometers in the region of interest. However, this is both impractical and potentially destructive. Instead, landscape ecology breaks the task into component parts relying on the Darwinian interdependence of environmental properties and fitness of specific species' phenotypic and genotypic properties. First, the satellite map is carefully analyzed to define the number and distribution of habitats. Then the species distribution in a representative sampling of each habitat is empirically determined. Ultimately, this permits sufficient bridging of spatial scales to accurately predict spatial distribution of plant and animal species within large regions.Currently, identifying intratumoral subpopulations requires detailed histological and molecular studies that are expensive and time consuming. Furthermore, this method is subject to sampling bias, is invasive for vital organs such as the brain, and inherently destructive precluding repeated assessments for monitoring post-treatment response and proteogenomic evolution. In contrast, modern cross-sectional imaging can interrogate the entire tumor noninvasively, allowing repeated analysis without disrupting the region of interest. In particular, magnetic resonance imaging (MRI) provides exceptional spatial resolution and generates signals that are unique to the molecular constituents of tissue. Here we propose that MRI scans may be the equivalent of satellite images in landscape ecology and, with appropriate application of Darwinian first principles and sophisticated image analytic methods, can be used to estimate regional variations in the molecular properties of cancer cells.We have initially examined this technique in glioblastoma, a malignant brain neoplasm which is morphologically complex and notorious for a fast progression from diagnosis to recurrence and death, making a suitable subject of noninvasive, rapidly repeated assessment of intratumoral evolution. Quantitative imaging analysis of routine clinical MRIs from glioblastoma has identified macroscopic morphologic characteristics which correlate with proteogenomics and prognosis. The key to the accurate detection and forecasting of intratumoral evolution using quantitative imaging analysis is likely to be in the understanding of the synergistic interactions between observable intratumoral subregions and the resulting tumor behavior.

Modern brain tumor imaging

The imaging and clinical management of patients with brain tumor continue to evolve over time and now heavily rely on physiologic imaging in addition to high-resolution structural imaging. Imaging remains a powerful noninvasive tool to positively impact the management of patients with brain tumor. This article provides an overview of the current state-of-the art clinical brain tumor imaging. In this review, we discuss general magnetic resonance (MR) imaging methods and their application to the diagnosis of, treatment planning and navigation, and disease monitoring in patients with brain tumor. We review the strengths, limitations, and pitfalls of structural imaging, diffusion-weighted imaging techniques, MR spectroscopy, perfusion imaging, positron emission tomography/MR, and functional imaging. Overall this review provides a basis for understudying the role of modern imaging in the care of brain tumor patients.

Evaluation of the microenvironmental heterogeneity in high-grade gliomas with IDH1/2 gene mutation using histogram analysis of diffusion-weighted imaging and dynamic-susceptibility contrast perfusion imaging

The purpose of our study was to explore the difference between isocitrate dehydrogenase (IDH)-1/2 gene mutation-positive and -negative high-grade gliomas (HGGs) using histogram analysis of apparent diffusion coefficient (ADC) and normalized cerebral blood volume (nCBV) maps. We enrolled 52 patients with histopathologically confirmed HGGs with IDH1/2 (P) (n = 16) or IDH1/2 (N) (n = 36). Histogram parameters of ADC and nCBV maps were correlated with gene mutations by using the unpaired student's t test and multivariable stepwise logistic regression analysis. The mean ADC value was higher in the IDH1 (P) group than IDH1 (N) (1,282.8 vs. 1,159.6 mm(2)/s, P = .0113). In terms of the cumulative ADC histograms, the 10th and 50th percentile values were also higher in the IDH1 (P) than IDH1 (N) (P = .0104 and .0183, respectively). We observed a higher 90th percentile value (3.121 vs. 2.397, P = .0208) and a steeper slope between the 10th (C10) and 90th (C90) of cumulative nCBV histograms (0.03386 vs. 0.02425/%, P = .0067) in the IDH1 (N) group. Multivariate analysis showed that the mean ADC mean value (P = .0048), the C90 value (P = .0113), and the slope between C10 and C90 (P = .0049) were the significant variables in the differentiation of IDH1 (P) from IDH1 (N). In conclusion, histogram analysis of ADC and nCBV maps based on entire tumor volume can be a useful tool for distinguishing IDH1 (P) and IDH1 (N), and it predicts that IDH (P) tumors have a more heterogeneous microenvironment than IDH (N) ones.

Gliomas: application of cumulative histogram analysis of normalized cerebral blood volume on 3 T MRI to tumor grading

Background

Glioma grading assumes significant importance in that low- and high-grade gliomas display different prognoses and are treated with dissimilar therapeutic strategies. The objective of our study was to retrospectively assess the usefulness of a cumulative normalized cerebral blood volume (nCBV) histogram for glioma grading based on 3 T MRI.

Methods

From February 2010 to April 2012, 63 patients with astrocytic tumors underwent 3 T MRI with dynamic susceptibility contrast perfusion-weighted imaging. Regions of interest containing the entire tumor volume were drawn on every section of the co-registered relative CBV (rCBV) maps and T2-weighted images. The percentile values from the cumulative nCBV histograms and the other histogram parameters were correlated with tumor grades. Cochran’s Q test and the McNemar test were used to compare the diagnostic accuracies of the histogram parameters after the receiver operating characteristic curve analysis. Using the parameter offering the highest diagnostic accuracy, a validation process was performed with an independent test set of nine patients.

Results

The 99th percentile of the cumulative nCBV histogram (nCBV C99), mean and peak height differed significantly between low- and high-grade gliomas (P = <0.001, 0.014 and <0.001, respectively) and between grade III and IV gliomas (P = <0.001, 0.001 and <0.001, respectively). The diagnostic accuracy of nCBV C99 was significantly higher than that of the mean nCBV (P = 0.016) in distinguishing high- from low-grade gliomas and was comparable to that of the peak height (P = 1.000). Validation using the two cutoff values of nCBV C99 achieved a diagnostic accuracy of 66.7% (6/9) for the separation of all three glioma grades.

Conclusion

Cumulative histogram analysis of nCBV using 3 T MRI can be a useful method for preoperative glioma grading. The nCBV C99 value is helpful in distinguishing high- from low-grade gliomas and grade IV from III gliomas.

MR imaging predictors of molecular profile and survival: multi-institutional study of the TCGA glioblastoma data set

PURPOSE

To conduct a comprehensive analysis of radiologist-made assessments of glioblastoma (GBM) tumor size and composition by using a community-developed controlled terminology of magnetic resonance (MR) imaging visual features as they relate to genetic alterations, gene expression class, and patient survival.

MATERIALS AND METHODS

Because all study patients had been previously deidentified by the Cancer Genome Atlas (TCGA), a publicly available data set that contains no linkage to patient identifiers and that is HIPAA compliant, no institutional review board approval was required. Presurgical MR images of 75 patients with GBM with genetic data in the TCGA portal were rated by three neuroradiologists for size, location, and tumor morphology by using a standardized feature set. Interrater agreements were analyzed by using the Krippendorff α statistic and intraclass correlation coefficient. Associations between survival, tumor size, and morphology were determined by using multivariate Cox regression models; associations between imaging features and genomics were studied by using the Fisher exact test.

RESULTS

Interrater analysis showed significant agreement in terms of contrast material enhancement, nonenhancement, necrosis, edema, and size variables. Contrast-enhanced tumor volume and longest axis length of tumor were strongly associated with poor survival (respectively, hazard ratio: 8.84, P = .0253, and hazard ratio: 1.02, P = .00973), even after adjusting for Karnofsky performance score (P = .0208). Proneural class GBM had significantly lower levels of contrast enhancement (P = .02) than other subtypes, while mesenchymal GBM showed lower levels of nonenhanced tumor (P < .01).

CONCLUSION

This analysis demonstrates a method for consistent image feature annotation capable of reproducibly characterizing brain tumors; this study shows that radiologists' estimations of macroscopic imaging features can be combined with genetic alterations and gene expression subtypes to provide deeper insight to the underlying biologic properties of GBM subsets.

SVM-based glioma grading: Optimization by feature reduction analysis

We investigated the predictive power of feature reduction analysis approaches in support vector machine (SVM)-based classification of glioma grade. In 101 untreated glioma patients, three analytic approaches were evaluated to derive an optimal reduction in features; (i) Pearson's correlation coefficients (PCC), (ii) principal component analysis (PCA) and (iii) independent component analysis (ICA). Tumor grading was performed using a previously reported SVM approach including whole-tumor cerebral blood volume (CBV) histograms and patient age. Best classification accuracy was found using PCA at 85% (sensitivity=89%, specificity=84%) when reducing the feature vector from 101 (100-bins rCBV histogram+age) to 3 principal components. In comparison, classification accuracy by PCC was 82% (89%, 77%, 2 dimensions) and 79% by ICA (87%, 75%, 9 dimensions). For improved speed (up to 30%) and simplicity, feature reduction by all three methods provided similar classification accuracy to literature values (∼87%) while reducing the number of features by up to 98%.