Quantitative combination of volumetric MR imaging and MR spectroscopy data for the discrimination of meningiomas from metastatic brain tumors by means of pattern recognition

Differentiation of multiple sclerosis lesions and low-grade brain tumors on MRS data: machine learning approaches

Some multiple sclerosis (MS) lesions may have great similarities with neoplastic brain lesions in magnetic resonance (MR) imaging and thus wrong diagnoses may occur. In this study, differentiation of MS and low-grade brain tumors was performed with computer-aided diagnosis (CAD) methods by magnetic resonance spectroscopy (MRS) data. MRS data belonging to 51 MS and 39 low-grade brain tumor patients were obtained. The feature extraction from MRS data was performed by the help of peak integration (PI) and full spectra (FS) methods and the most significant features were identified. For the classification step, artificial neural network (ANN), support vector machine (SVM), and linear discriminant analysis (LDA) methods were used and the differentiation between MS and brain tumor was performed automatically. Examining the results, one can conclude that data which belong to MS and low-grade brain tumor cases were automatically differentiated from each other with the help of ANN with 100% accuracy, 100% sensitivity, and 100% specificity. Using of MR spectroscopy and artificial intelligence methods may be useful as a complementary imaging technique to MR imaging in the differentiation of MS lesions and low-grade brain tumors.

Differentiation of relapsing-remitting and secondary progressive multiple sclerosis: a magnetic resonance spectroscopy study based on machine learning

INTRODUCTION

Magnetic resonance imaging (MRI) is the most important tool for diagnosis and follow-up in multiple sclerosis (MS). The discrimination of relapsing-remitting MS (RRMS) from secondary progressive MS (SPMS) is clinically difficult, and developing the proposal presented in this study would contribute to the process.

OBJECTIVE

This study aimed to ensure the automatic classification of healthy controls, RRMS, and SPMS by using MR spectroscopy and machine learning methods.

METHODS

MR spectroscopy (MRS) was performed on a total of 91 participants, distributed into healthy controls (n=30), RRMS (n=36), and SPMS (n=25). Firstly, MRS metabolites were identified using signal processing techniques. Secondly, feature extraction was performed based on MRS Spectra. N-acetylaspartate (NAA) was the most significant metabolite in differentiating MS types. Lastly, binary classifications (healthy controls-RRMS and RRMS-SPMS) were carried out according to features obtained by the Support Vector Machine algorithm.

RESULTS

RRMS cases were differentiated from healthy controls with 85% accuracy, 90.91% sensitivity, and 77.78% specificity. RRMS and SPMS were classified with 83.33% accuracy, 81.81% sensitivity, and 85.71% specificity.

CONCLUSIONS

A combined analysis of MRS and computer-aided diagnosis may be useful as a complementary imaging technique to determine MS types.

Application of MR morphologic, diffusion tensor, and perfusion imaging in the classification of brain tumors using machine learning scheme

PURPOSE

While MRI is the modality of choice for the assessment of patients with brain tumors, differentiation between various tumors based on their imaging characteristics might be challenging due to overlapping imaging features. The purpose of this study was to apply a machine learning scheme using basic and advanced MR sequences for distinguishing different types of brain tumors.

METHODS

The study cohort included 141 patients (41 glioblastoma, 38 metastasis, 50 meningioma, and 12 primary central nervous system lymphoma). A computer-assisted classification scheme, combining morphologic MRI, perfusion MRI, and DTI metrics, was developed and used for tumor classification. The proposed multistep scheme consists of pre-processing, ROI definition, features extraction, feature selection, and classification. Feature subset selection was performed using support vector machines (SVMs). Classification performance was assessed by leave-one-out cross-validation. Given an ROI, the entire classification process was done automatically via computer and without any human intervention.

RESULTS

A binary hierarchical classification tree was chosen. In the first step, selected features were chosen for distinguishing glioblastoma from the remaining three classes, followed by separation of meningioma from metastasis and PCNSL, and then to discriminate PCNSL from metastasis. The binary SVM classification accuracy, sensitivity and specificity for glioblastoma, metastasis, meningiomas, and primary central nervous system lymphoma were 95.7, 81.6, and 91.2%; 92.7, 95.1, and 93.6%; 97, 90.8, and 58.3%; and 91.5, 90, and 96.9%, respectively.

CONCLUSION

A machine learning scheme using data from anatomical and advanced MRI sequences resulted in high-performance automatic tumor classification algorithm. Such a scheme can be integrated into clinical decision support systems to optimize tumor classification.

Preoperative grading of intracranial meningioma by magnetic resonance spectroscopy (1H-MRS)

Although proton magnetic resonance spectroscopy (¹H-MRS) is a common method for the evaluation of intracranial meningiomas, controversy exists regarding which parameter of ¹H-MRS best predicts the histopathological grade of an intracranial meningioma. In this study, we evaluated the results of pre-operative ¹H-MRS to identify predictive factors for high-grade intracranial meningioma. Thirteen patients with World Health Organization (WHO) grade II-III meningioma (confirmed by pathology) were defined as high-grade; twenty-two patients with WHO grade I meningioma were defined as low-grade. All patients were evaluated by ¹H-MRS before surgery. The relationships between the ratios of metabolites (N-acetylaspartate [NAA], creatine [Cr], and choline [Cho]) and the diagnosis of high-grade meningioma were analyzed. According to Mann-Whitney U test analysis, the Cho/NAA ratio in cases of high-grade meningioma was significantly higher than in cases of low-grade meningioma (6.34 ± 7.90 vs. 1.58 ± 0.77, p<0.05); however, there were no differences in age, Cho/Cr, or NAA/Cr. According to conditional inference tree analysis, the optimal cut-off point for the Cho/NAA ration between high-grade and low-grade meningioma was 2.409 (sensitivity = 61.54%; specificity = 86.36%). This analysis of pre-operative ¹H-MRS metabolite ratio demonstrated that the Cho/NAA ratio may provide a simple and practical predictive value for high-grade intracranial meningiomas, and may aid neurosurgeons in efforts to design an appropriate surgical plan and treatment strategy before surgery.

Radiomic Phenotyping in Brain Cancer to Unravel Hidden Information in Medical Images

Radiomics is a new area of research in the field of imaging with tremendous potential to unravel the hidden information in digital images. The scope of radiology has grown exponentially over the last two decades; since the advent of radiomics, many quantitative imaging features can now be extracted from medical images through high-throughput computing, and these can be converted into mineable data that can help in linking imaging phenotypes with clinical data, genomics, proteomics, and other "omics" information. In cancer, radiomic imaging analysis aims at extracting imaging features embedded in the imaging data, which can act as a guide in the disease or cancer diagnosis, staging and planning interventions for treating patients, monitor patients on therapy, predict treatment response, and determine patient outcomes.

Metabolic Imaging in Humans

Metabolic imaging enhances understanding of disease metabolisms and holds great potential as a measurement tool for evaluating disease prognosis and treatment effectiveness. Advancement of techniques, such as magnetic resonance spectroscopy, positron emission tomography, and mass spectrometry, allows for improved accuracy for quantification of metabolites and present unique possibilities for use in clinic. This article reviews and discusses literature reports of metabolic imaging in humans published since 2010 according to disease type, including cancer, degenerative disorders, psychiatric disorders, and others, as well as the current application of the various related techniques.

Multidimensional texture characterization: on analysis for brain tumor tissues using MRS and MRI

This paper investigates the efficacy of automated pattern recognition methods on magnetic resonance data with the objective of assisting radiologists in the clinical diagnosis of brain tissue tumors. In this paper, the sciences of magnetic resonance imaging (MRI) and magnetic resonance spectroscopy (MRS) are combined to improve the accuracy of the classifier, based on the multidimensional co-occurrence matrices to assess the detection of pathological tissues (tumor and edema), normal tissues (white matter - WM and gray matter - GM), and fluid (cerebrospinal fluid - CSF). The results show the ability of the classifier with iterative training to automatically and simultaneously recover tissue-specific spectral and structural patterns and achieve segmentation of tumor and edema and grading of high and low glioma tumor. Here, extreme learning machine - improved particle swarm optimization (ELM-IPSO) neural network classifier is trained with the feature descriptions in brain magnetic resonance (MR) spectra. This has the characteristics of varying the normal spectral pattern associated with tumor patterns along with imaging features. Validation was performed considering 35 clinical studies. The volumetric features extracted from the vectors of this matrix articulate some important elementary structures, which along with spectroscopic metabolite ratios discriminate the tumor grades and tissue classes. The quantitative 3D analysis reveals significant improvement in terms of global accuracy rate for automatic classification in brain tissues and discriminating pathological tumor tissue from structural healthy brain tissue.

Quantification of heterogeneity as a biomarker in tumor imaging: a systematic review

BACKGROUND

Many techniques are proposed for the quantification of tumor heterogeneity as an imaging biomarker for differentiation between tumor types, tumor grading, response monitoring and outcome prediction. However, in clinical practice these methods are barely used. This study evaluates the reported performance of the described methods and identifies barriers to their implementation in clinical practice.

METHODOLOGY

The Ovid, Embase, and Cochrane Central databases were searched up to 20 September 2013. Heterogeneity analysis methods were classified into four categories, i.e., non-spatial methods (NSM), spatial grey level methods (SGLM), fractal analysis (FA) methods, and filters and transforms (F&T). The performance of the different methods was compared.

PRINCIPAL FINDINGS

Of the 7351 potentially relevant publications, 209 were included. Of these studies, 58% reported the use of NSM, 49% SGLM, 10% FA, and 28% F&T. Differentiation between tumor types, tumor grading and/or outcome prediction was the goal in 87% of the studies. Overall, the reported area under the curve (AUC) ranged from 0.5 to 1 (median 0.87). No relation was found between the performance and the quantification methods used, or between the performance and the imaging modality. A negative correlation was found between the tumor-feature ratio and the AUC, which is presumably caused by overfitting in small datasets. Cross-validation was reported in 63% of the classification studies. Retrospective analyses were conducted in 57% of the studies without a clear description.

CONCLUSIONS

In a research setting, heterogeneity quantification methods can differentiate between tumor types, grade tumors, and predict outcome and monitor treatment effects. To translate these methods to clinical practice, more prospective studies are required that use external datasets for validation: these datasets should be made available to the community to facilitate the development of new and improved methods.

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.

The role of diffusion and perfusion weighted imaging in the differential diagnosis of cerebral tumors: a review and future perspectives

The role of conventional Magnetic Resonance Imaging (MRI) in the detection of cerebral tumors has been well established. However its excellent soft tissue visualization and variety of imaging sequences are in many cases non-specific for the assessment of brain tumor grading. Hence, advanced MRI techniques, like Diffusion-Weighted Imaging (DWI), Diffusion Tensor Imaging (DTI) and Dynamic-Susceptibility Contrast Imaging (DSCI), which are based on different contrast principles, have been used in the clinical routine to improve diagnostic accuracy. The variety of quantitative information derived from these techniques provides significant structural and functional information in a cellular level, highlighting aspects of the underlying brain pathophysiology. The present work, reviews physical principles and recent results obtained using DWI/DTI and DSCI, in tumor characterization and grading of the most common cerebral neoplasms, and discusses how the available MR quantitative data can be utilized through advanced methods of analysis, in order to optimize clinical decision making.

Comparison of sagittal and transverse echo planar spectroscopic imaging on the quantification of brain metabolites

PURPOSE

We quantitatively compared sagittal and transverse echo planar spectroscopic imaging (EPSI) on the quantification of metabolite concentrations with consideration of tissue variation. A quantification strategy is proposed to collect the necessary information for quantification of concentrations in a minimized acquisition time.

METHODS

Six transverse and six sagittal EPSI data were collected on healthy volunteers. Metabolite concentrations of N-acetyl-aspartate (NAA), total creatine (tCr), total choline (tCho), myo-inositol (mI), and glutamate and glutamine complex (Glx) were quantified using water scaling with partial volume and relaxation correction. Linear regression analysis was performed to extract concentrations in gray matter (GM) and white matter (WM). The inter- and intrasubject coefficients of variance (CV) were estimated.

RESULTS

Concentrations and fitting errors of sagittal and transverse EPSI were at same level. GM to WM contrast of concentrations was found in NAA, tCr, and tCho. The intersubject CVs revealed greater variability in the sagittal EPSI than in the transverse EPSI. The intrasubject CVs of the transverse EPSI were below 5% for NAA, tCr, and tCho.

CONCLUSION

We showed that quantified concentrations of sagittal and transverse EPSI after partial volume correction are comparable and reproducible. The proposed quantification strategy can be conveniently adapted into various MRI protocols.

Differential diagnosis of intracranial meningiomas based on magnetic resonance spectroscopy

BACKGROUND AND PURPOSE

To determine in vivo magnetic resonance spectroscopy (MRS) characteristics of intracranial meningiomas and to assess MRS reliability in meningioma grading and discrimination from tumours of similar radiological appearance, such as lymphomas, schwannomas and haemangiopericytomas.

MATERIAL AND METHODS

Analysis of spectra of 14 patients with meningiomas, 6 with schwannomas, 2 with lymphomas, 2 with haemangiopericytomas and 17 control spectra taken from healthy hemispheres.

RESULTS

All the patients with meningiomas had a high Cho signal (long TE). There were very low signals of Naa and Cr in the spectra of 10 patients. A reversed Ala doublet was seen only in 2 cases. Four patients had a negative Lac signal, whereas 3 had high Lac-Lip spectra. Twelve spectra showed high Cho signals (short TE). In one case the Cho signal was extremely low. All spectra displayed a very low Cr signal, but high Glx and Lac-Lip signals. Ala presence was found only in 3 patients. The mean Cho/Cr ratio (PRESS) was 5.97 (1.12 in normal brain, p < 0.05). Lac-Lip was present in all the meningiomas (STEAM). The Ala signal was seen only in 2 spectra with long TE and in 3 sequences of the short TE sequences. There were both β/γ-Glx and α-Glx/glutathione signals in all 14 meningiomas.

CONCLUSIONS

MRS is unable to discriminate low and high grade meningiomas. The method seems to be helpful in discriminating lymphomas (absent Glx signal), schwannomas (mI signal in the short TE sequences) and haemangiopericytomas (presence of mI band) from meningiomas.

A survey of MRI-based medical image analysis for brain tumor studies

MRI-based medical image analysis for brain tumor studies is gaining attention in recent times due to an increased need for efficient and objective evaluation of large amounts of data. While the pioneering approaches applying automated methods for the analysis of brain tumor images date back almost two decades, the current methods are becoming more mature and coming closer to routine clinical application. This review aims to provide a comprehensive overview by giving a brief introduction to brain tumors and imaging of brain tumors first. Then, we review the state of the art in segmentation, registration and modeling related to tumor-bearing brain images with a focus on gliomas. The objective in the segmentation is outlining the tumor including its sub-compartments and surrounding tissues, while the main challenge in registration and modeling is the handling of morphological changes caused by the tumor. The qualities of different approaches are discussed with a focus on methods that can be applied on standard clinical imaging protocols. Finally, a critical assessment of the current state is performed and future developments and trends are addressed, giving special attention to recent developments in radiological tumor assessment guidelines.