Three-dimensional texture analysis of MRI brain datasets

Kovalev, Vassili; Kruggel, Frithjof; Gertz, H.-J.; Cramon, D. Yves von

doi:10.1109/42.925295

Cited by 162 publications

(123 citation statements)

References 23 publications

Supporting

Mentioning

121

Contrasting

Unclassified

Order By: Relevance

“…We used extended multisort cooccurrence matrices (Kovalev et al, 2001) as detailed descriptors of the spatial image structure. These descriptors are computed from elementary features of voxel pairs within a subregion, such as intensity, intensity gradient magnitude (local intensity variation), the angle between gradient vectors (spatial coherence of intensity slopes), and the intervoxel distance.…”

Section: Cooccurrence Image Descriptorsmentioning

confidence: 99%

“…Gradient magnitudes G(i) and G(k) were calculated from the orthogonal intensity gradient vector components G x , G y and G z as elements (which is proportional to the brain region volume) for each distance bin separately. Refer to Kovalev et al, 2001 for a more detailed description. Intensity, gradient magnitude, and relative gradient orientation were chosen as cooccurrence matrix dimensions because they form a complementary set of elementary image features.…”

Section: K) Let Us Denote Intensities Of These Voxels By I(i) and I(mentioning

confidence: 99%

“…Explicit tissue classification is avoided by using 3D image descriptors that properly describe the complex spatial gray-scale structure of T 1 -weighted MRI brain datasets. Here, we employed image texture measures as captured by extended multisort cooccurrence matrices (Kovalev et al, 2001).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Gender and age effects in structural brain asymmetry as measured by MRI texture analysis

Kovalev¹,

Kruggel²,

Cramon³

2003

NeuroImage

140

View full text Add to dashboard Cite

Effects of gender and age on structural brain asymmetry were studied by 3D texture analysis in 380 adults. Asymmetry is detected by comparing the complex 3D gray-scale image patterns in the left and right cerebral hemispheres as revealed by anatomical T1-weighted MRI datasets. The Talairach and Tournoux parcellation system was applied to study the asymmetry on five levels: the whole cerebrum, nine coronal sections, 12 axial sections, boxes resulting from both coronal and axial subdivisions, and by a sliding spherical window of 9 mm diameter. The analysis revealed that the brain asymmetry increases in the anterior-posterior direction starting from the central region onward. Male brains were found to be more asymmetric than female. This gender-related effect is noticeable in all brain areas but is most significant in the superior temporal gyrus, Heschl's gyrus, the adjacent white matter regions in the temporal stem and the knee of the optic radiation, the thalamus, and the posterior cingulate. The brain asymmetry increases significantly with age in the inferior frontal gyrus, anterior insula, anterior cingulate, parahippocampal gyrus, retrosplenial cortex, coronal radiata, and knee region of the internal capsule. Asymmetry decreases with age in the optic radiation, precentral gyrus, and angular gyrus. The texture-based method reported here is based on extended multisort cooccurrence matrices that employ intensity, gradient, and anisotropy features in a uniform way. It is sensitive, simple to reproduce, robust, and unbiased in the sense that segmentation of brain compartments and spatial transformations are not necessary. Thus, it should be considered as another tool for digital morphometry in neuroscience.

show abstract

Section: Cooccurrence Image Descriptorsmentioning

confidence: 99%

Section: K) Let Us Denote Intensities Of These Voxels By I(i) and I(mentioning

confidence: 99%

See 1 more Smart Citation

Gender and age effects in structural brain asymmetry as measured by MRI texture analysis

Kovalev¹,

Kruggel²,

Cramon³

2003

NeuroImage

140

View full text Add to dashboard Cite

show abstract

“…This process would allow for the extraction of more useful underlying information based on quantitatively derived features: Radiomics. Several institutes have reported quantitative analysis studies, with a focus on radiomic features, for different imaging modalities such as computed tomography (CT),1, 2, 3 and magnetic resonance imaging (MRI) 4, 5, 6. The investigation of positron emission tomography (PET) radiomics was first reported in 2009 7, 8, 9.…”

Section: Introductionmentioning

confidence: 99%

Reproducibility of F18‐FDG PET radiomic features for different cervical tumor segmentation methods, gray‐level discretization, and reconstruction algorithms

Altazi

Zhang

Fernandez

et al. 2017

J Applied Clin Med Phys

View full text Add to dashboard Cite

Site‐specific investigations of the role of radiomics in cancer diagnosis and therapy are emerging. We evaluated the reproducibility of radiomic features extracted from 18Flourine–fluorodeoxyglucose (18F‐FDG) PET images for three parameters: manual versus computer‐aided segmentation methods, gray‐level discretization, and PET image reconstruction algorithms. Our cohort consisted of pretreatment PET/CT scans from 88 cervical cancer patients. Two board‐certified radiation oncologists manually segmented the metabolic tumor volume (MTV1 and MTV2) for each patient. For comparison, we used a graphical‐based method to generate semiautomated segmented volumes (GBSV). To address any perturbations in radiomic feature values, we down‐sampled the tumor volumes into three gray‐levels: 32, 64, and 128 from the original gray‐level of 256. Finally, we analyzed the effect on radiomic features on PET images of eight patients due to four PET 3D‐reconstruction algorithms: maximum likelihood‐ordered subset expectation maximization (OSEM) iterative reconstruction (IR) method, fourier rebinning‐ML‐OSEM (FOREIR), FORE‐filtered back projection (FOREFBP), and 3D‐Reprojection (3DRP) analytical method. We extracted 79 features from all segmentation method, gray‐levels of down‐sampled volumes, and PET reconstruction algorithms. The features were extracted using gray‐level co‐occurrence matrices (GLCM), gray‐level size zone matrices (GLSZM), gray‐level run‐length matrices (GLRLM), neighborhood gray‐tone difference matrices (NGTDM), shape‐based features (SF), and intensity histogram features (IHF). We computed the Dice coefficient between each MTV and GBSV to measure segmentation accuracy. Coefficient values close to one indicate high agreement, and values close to zero indicate low agreement. We evaluated the effect on radiomic features by calculating the mean percentage differences (d¯) between feature values measured from each pair of parameter elements (i.e. segmentation methods: MTV1‐MTV2, MTV1‐GBSV, MTV2‐GBSV; gray‐levels: 64‐32, 64‐128, and 64‐256; reconstruction algorithms: OSEM‐FORE‐OSEM, OSEM‐FOREFBP, and OSEM‐3DRP). We used false|normald¯false| as a measure of radiomic feature reproducibility level, where any feature scored false|normald¯false| ±SD ≤ |25|% ± 35% was considered reproducible. We used Bland–Altman analysis to evaluate the mean, standard deviation (SD), and upper/lower reproducibility limits (U/LRL) for radiomic features in response to variation in each testing parameter. Furthermore, we proposed U/LRL as a method to classify the level of reproducibility: High— ±1% ≤ U/LRL ≤ ±30%; Intermediate— ±30% < U/LRL ≤ ±45%; Low— ±45 < U/LRL ≤ ±50%. We considered any feature below the low level as nonreproducible (NR). Finally, we calculated the interclass correlation coefficient (ICC) to evaluate the reliability of radiomic feature measurements for each parameter. The segmented volumes of 65 patients (81.3%) scored Dice coefficient >0.75 for all three volumes. The result outcomes revealed a tendency of higher radiomic fe...

show abstract

“…Mahmoud et al [22] have proposed a 3D approach using co-occurrence matrix analysis to increase the sensitivity and specificity of brain tumor characterization with promising results. Kovalev et al [15][16][17] tested 3D co-occurrence matrices in analyzing cerebral tissue and glioma in T1-weighted MR images and analysis in age/ gender related differences.…”

Section: Volume Of Interest Extraction and Feature Calculationmentioning

confidence: 99%

Multidimensional Texture Characterization: On Analysis for Brain Tumor Tissues Using MRS and MRI

Nachimuthu¹,

Devi²

2014

J Digit Imaging

View full text Add to dashboard Cite

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.

show abstract

Three-dimensional texture analysis of MRI brain datasets

Cited by 162 publications

References 23 publications

Gender and age effects in structural brain asymmetry as measured by MRI texture analysis

Gender and age effects in structural brain asymmetry as measured by MRI texture analysis

Reproducibility of F18‐FDG PET radiomic features for different cervical tumor segmentation methods, gray‐level discretization, and reconstruction algorithms

Multidimensional Texture Characterization: On Analysis for Brain Tumor Tissues Using MRS and MRI

Contact Info

Product

Resources

About