The objective of this study is to investigate the use of pattern classification methods for distinguishing different types of brain tumors, such as primary gliomas from metastases, and also for grading of gliomas. The availability of an automated computer analysis tool that is more objective than human readers can potentially lead to more reliable and reproducible brain tumor diagnostic procedures. A computer-assisted classification method combining conventional MRI and perfusion MRI is developed and used for differential diagnosis. The proposed scheme consists of several steps including region-of-interest definition, feature extraction, feature selection, and classification. The extracted features include tumor shape and intensity characteristics, as well as rotation invariant texture features. Feature subset selection is performed using support vector machines with recursive feature elimination. The method was applied on a population of 102 brain tumors histologically diagnosed as metastasis (24), meningiomas (4), gliomas World Health Organization grade II (22), gliomas World Health Organization grade III (18), and glioblastomas (34). The binary support vector machine classification accuracy, sensitivity, and specificity, assessed by leave-one-out cross-validation, were, respectively, 85%, 87%, and 79% for discrimination of metastases from gliomas and 88%, 85%, and 96% for discrimination of high-grade (grades III and IV) from low-grade (grade II) neoplasms. Multiclass classification was also performed via a onevs-all voting scheme.

BACKGROUND AND PURPOSE: Glioblastomas, brain metastases, and PCLs may have similar enhancement patterns on MR imaging, making the differential diagnosis difficult or even impossible. The purpose of this study was to determine whether a combination of DTI and DSC can assist in the differentiation of glioblastomas, solitary brain metastases, and PCLs.

OBJECTIVE The objective of our study was to predict response to chemoradiation therapy in patients with head and neck squamous cell carcinoma (HNSCC) by combined use of diffusion-weighted imaging (DWI) and high-spatial-resolution, high-temporal-resolution dynamic contrast-enhanced MRI (DCE-MRI) parameters from primary tumors and metastatic nodes. SUBJECTS AND METHODS Thirty-two patients underwent pretreatment DWI and DCE-MRI using a modified radial imaging sequence. Postprocessing of data included motion-correction algorithms to reduce motion artifacts. The median apparent diffusion coefficient (ADC), volume transfer constant (Ktrans), extracellular extravascular volume fraction (ve), and plasma volume fraction (vp) were computed from primary tumors and nodal masses. The quality of the DCE-MRI maps was estimated using a threshold median chi-square value of 0.10 or less. Multivariate logistic regression and receiver operating characteristic curve analyses were used to determine the best model to discriminate responders from nonresponders. RESULTS Acceptable χ2 values were observed from 84% of primary tumors and 100% of nodal masses. Five patients with unsatisfactory DCE-MRI data were excluded and DCEMRI data for three patients who died of unrelated causes were censored from analysis. The median follow-up for the remaining patients (n = 24) was 23.72 months. When ADC and DCE-MRI parameters (Ktrans, ve, vp) from both primary tumors and nodal masses were incorporated into multivariate logistic regression analyses, a considerably higher discriminative accuracy (area under the curve [AUC] = 0.85) with a sensitivity of 81.3% and specificity of 75% was observed in differentiating responders (n = 16) from nonresponders (n = 8). CONCLUSION The combined use of DWI and DCE-MRI parameters from both primary tumors and nodal masses may aid in prediction of response to chemoradiation therapy in patients with HNSCC.

MR-visible lipids or mobile lipids are defined as lipids that are observable using proton magnetic resonance spectroscopy in cells and in tissues. These MR-visible lipids are composed of triglycerides and cholesterol esters that accumulate in intracellular neutral lipid droplets, where their MR visibility is conferred as a result of the increased molecular motion available in this unique physical environment. This review will discuss factors that lead to the biogenesis of MR-visible lipids in cancer cells and in other cell types such as immune cells and fibroblasts. We focus on the accumulations of mobile lipids that are inducible in cultured cells by a number of stresses, including culture conditions and in response to activating stimuli or apoptotic cell death induced by anticancer drugs. This is compared with animal tumor models, where increases in mobile lipids are observed in response to chemo and radiotherapy, and to human tumors where mobile lipids are observed predominantly in high-grade brain tumors and in regions of necrosis. Conducive conditions for mobile lipid formation in the tumor microenvironment will be discussed including low pH, oxygen availability and the presence of inflammatory cells. It is concluded that MR-visible lipids appear in cancer cells and human tumors as a stress response. Mobile lipids stored as neutral lipid droplets may play a role in detoxification of the cell or act as an alternate energy source, especially in cancer cells, which often grow in ischemic/hypoxic environments. The role of MR-visible lipids in cancer diagnosis and assessment of treatment response both in animal models of cancer as well as human brain tumors will also be discussed. Although technical limitations exist in the accurate detection of intratumoral mobile lipids, early increases in mobile lipids after therapeutic interventions may be used as a potential biomarker for assessing treatment response in cancer.

The purpose of this study is to determine whether diffusion tensor imaging (DTI) metrics including tensor shape measures such as linear and planar anisotropy coefficients (CL and CP) can help differentiate glioblastomas from solitary brain metastases. Sixty-three patients with histopathologic diagnosis of glioblastomas (22 men, 16 women, mean age 58.4 years) and brain metastases (13 men, 12 women, mean age 56.3 years) were included in this study. Contrast-enhanced T1-weighted, fluid attenuated inversion recovery (FLAIR) images, fractional anisotropy (FA), apparent diffusion coefficient (ADC), CL and CP maps were co-registered and each lesion was semi-automatically subdivided into four regions: central, enhancing, immediate peritumoral and distant peritumoral. DTI metrics as well as the normalized signal intensity from the contrast-enhanced T1-weighted images were measured from each region. Univariate and multivariate logistic regression analyses were employed to determine the best model for classification. The results demonstrated that FA, CL and CP from glioblastomas were significantly higher than those of brain metastases from all segmented regions (p < 0.05), and the differences from the enhancing regions were most significant (p < 0.001). FA and CL from the enhancing region had the highest prediction accuracy when used alone with an area under the curve of 0.90. The best logistic regression model included three parameters (ADC, FA and CP) from the enhancing part, resulting in 92% sensitivity, 100% specificity and area under the curve of 0.98. We conclude that DTI metrics, used individually or combined, have a potential as a noninvasive measure to differentiate glioblastomas from metastases.