Gray-level Co-occurrence Matrix Entropy Research Articles

Baseline Computed Tomography Radiomic and Genomic Assessment of Head and Neck Squamous Cell Carcinoma.

To determine the relationship between computed tomography (CT) radiomic features and gene expression levels in head and neck squamous cell carcinoma (HNSCC). This retrospective study included 66 patients with HNSCC primary lesions (36 oropharyngeal, 6 hypopharyngeal, 10 laryngeal, 14 oral cavity). Gene expression information for 6 targetable genes (fibroblast growth factor receptor [FGFR]1, epidermal growth factor receptor [EGFR], FGFR2, FGFR3, EPHA2, PIK3CA) was obtained via Agilent microarrays from samples collected between 1997 and 2010. Pretreatment contrast-enhanced soft tissue neck CT scans were reviewed, and 142 radiomics features were derived. R was used to calculate Pearson correlation coefficients were calculated between gene expression levels and each radiomic feature. P values were adjusted using the false discovery rate (FDR) method. There were significant correlations between FGFR1 and 5 gray level cooccurrence matrix (GLCM) features with FDR-adjusted P values less than 0.05: inertia (r = 0.366, FDR-adjusted P = 0.006), absolute value (r = 0.31, FDR-adjusted P = 0.024), contrast (r = 0.366, FDR-adjusted P = 0.006), difference average (r = 0.31, FDR-adjusted P = 0.024), and difference variance (r = 0.37, FDR-adjusted P = 0.005). There was 1 correlated feature for FGFR2 with an FDR-adjusted P value less than 0.05: fractal dimension box-coarse (r = 0.33, FDR-adjusted P = 0.018). There was 1 correlated feature for EPHA2 with an FDR-adjusted P value less than 0.05: GLCM entropy (r = -0.28, FDR-adjusted P = 0.049). Six of the 7 features that showed significant correlation belonged to the GLCM class of features. The CT radiomic features demonstrate correlations with FGFR1 status in HNSCC and should be further investigated for their potential to predict FGFR1 status.

Application of computerized 3D-CT texture analysis of pancreas for the assessment of patients with diabetes.

ObjectiveTo evaluate the role of computerized 3D CT texture analysis of the pancreas as quantitative parameters for assessing diabetes.MethodsAmong 2,493 patients with diabetes, 39 with type 2 diabetes (T2D) and 12 with type 1 diabetes (T1D) who underwent CT using two selected CT scanners, were enrolled. We compared these patients with age-, body mass index- (BMI), and CT scanner-matched normal subjects. Computerized texture analysis for entire pancreas was performed by extracting 17 variable features. A multivariate logistic regression analysis was performed to identify the predictive factors for diabetes. A receiver operator characteristic (ROC) curve was constructed to determine the optimal cut off values for statistically significant variables.ResultsIn diabetes, mean attenuation, standard deviation, variance, entropy, homogeneity, surface area, sphericity, discrete compactness, gray-level co-occurrence matrix (GLCM) contrast, and GLCM entropy showed significant differences (P < .05). Multivariate analysis revealed that a higher variance (adjusted OR, 1.002; P = .005), sphericity (adjusted OR, 1.649×104; P = .048), GLCM entropy (adjusted OR, 1.057×105; P = .032), and lower GLCM contrast (adjusted OR, 0.997; P < .001) were significant variables. The mean AUCs for each feature were 0.654, 0.689, 0.620, and 0.613, respectively (P < .05). In subgroup analysis, only larger surface area (adjusted OR, 1.000; P = .025) was a significant predictor for T2D.ConclusionsComputerized 3D CT texture analysis of the pancreas could be helpful for predicting diabetes. A higher variance, sphericity, GLCM entropy, and a lower GLCM contrast were the significant predictors for diabetes.

103P - Contrast-enhanced CT-based textural parameters as potential prognostic factors of survival for colorectal cancer patients receiving targeted therapy

BackgroundTexture analysis is used to assess the homogeneous or nonhomogenous information of the selected regions, and provides many quantitative and semiquantitative parameters that are able to closely reflect the characteristics of cancer lesions. This study was designed to estimate the clinical significance of the contrast-enhanced computed tomograpgy (CT) textural features for prediction of survival in colorectal cancer (CRC) patients receiving targeted therapy (bevacizumab and cetuximab). MethodsThe LifeX software was used to extract the textural parameters of the tumor lesions in the contrast-enhanced CT. Univariate and multivariate analyses using the Cox proportional hazards model were performed to assess the prognostic value of textural parameters. ResultsEighty CRC patients receiving targeted therapy (bevacizumab 42; cetuximab 38) were included. In the multivariate analysis, 8 textural parameters were revealed to be independent predictors (p < 0.05) of progression-free survival (PFS) and overall survival (OS), including Skewness, Kurtosis, Homogeneity, Energy and Entropy of Gray-level co-occurrence matrix (GLCM), Long-run emphasis (LRE), Long-run high gray-level emphasis (LRHGE), and Contrast. Furthermore, Sphericity, Compacity, Long-run low gray-level emphasis (LRLGE), Low gray-level zone emphasis (LGZE) and Short-zone low gray-level emphasis (SZLGE) were show to be significantly associated with PFS, while Entropy-Histogram and Energy-Histogram, Dissimilarty, Short-run emphasis (SRE), Short-run low gray-level emphasis (SRLGE), Run percentage (RP), Long-zone emphasis (LZE), Long-zone low gray-level emphasis (LZLGE) and Long-zone high gray-level emphasis (LZHGE) were significantly associated with OS. ConclusionsIn conclusion, our study provides preliminary evidences that several radiomic parameters of tumor lesions derived from CT images were prognostic factors and predictive markers for CRC patients who are candidates for targeted therapy (bevacizumab and cetuximab). Legal entity responsible for the studyThe authors. FundingHas not received any funding. DisclosureAll authors have declared no conflicts of interest.

Prognostic Value of Functional Parameters of 18F-FDG-PET Images in Patients with Primary Renal/Adrenal Lymphoma.

Objectives The aim of this study is to explore the textural features that may identify the morphological changes in the lymphoma region and predict the prognosis of patients with primary renal lymphoma (PRL) and primary adrenal lymphoma (PAL). Methods This retrospective study comprised nineteen non-Hodgkin's lymphoma (NHL) patients undergoing 18F-FDG-PET/CT at West China Hospital from December 2013 to May 2017. 18F-FDG-PET images were reviewed independently by two board certificated radiologists of nuclear medicine, and the texture features were extracted from LifeX packages. The prognostic value of PET FDG-uptake parameters, patients' baseline characteristics, and textural parameters were analyzed using Kaplan–Meier analysis. Cox regression analysis was used to identify the independent prognostic factors among the imaging and clinical features. Results The overall survival of included patients was 18.84 ± 13.40 (mean ± SD) months. Univariate Cox analyses found that the tumor stage, GLCM (gray-level co-occurrence matrix) entropy, GLZLM_GLNU (gray-level nonuniformity), and GLZLM_ZLNU (zone length nonuniformity), values were significant predictors for OS. Among them, GLRLM_RLNU ≥216.6 demonstrated association with worse OS at multivariate analysis (HR 9.016, 95% CI 1.041–78.112, p=0.046). Conclusions The texture analysis of 18F-FDG-PET images could potentially serve as a noninvasive strategy to predict the overall survival of patients with PRL and PAL.

Prediction of Chemotherapy Response of Osteosarcoma Using Baseline 18F-FDG Textural Features Machine Learning Approaches with PCA

Purpose Patients with high-grade osteosarcoma undergo several chemotherapy cycles before surgical intervention. Response to chemotherapy, however, is affected by intratumor heterogeneity. In this study, we assessed the ability of a machine learning approach using baseline 18F-fluorodeoxyglucose (18F-FDG) positron emitted tomography (PET) textural features to predict response to chemotherapy in osteosarcoma patients. Materials and Methods This study included 70 osteosarcoma patients who received neoadjuvant chemotherapy. Quantitative characteristics of the tumors were evaluated by standard uptake value (SUV), total lesion glycolysis (TLG), and metabolic tumor volume (MTV). Tumor heterogeneity was evaluated using textural analysis of 18F-FDG PET scan images. Assessments were performed at baseline and after chemotherapy using 18F-FDG PET; 18F-FDG textural features were evaluated using the Chang-Gung Image Texture Analysis toolbox. To predict the chemotherapy response, several features were chosen using the principal component analysis (PCA) feature selection method. Machine learning was performed using linear support vector machine (SVM), random forest, and gradient boost methods. The ability to predict chemotherapy response was evaluated using the area under the receiver operating characteristic curve (AUC). Results AUCs of the baseline 18F-FDG features SUVmax, TLG, MTV, 1st entropy, and gray level co-occurrence matrix entropy were 0.553, 0538, 0.536, 0.538, and 0.543, respectively. However, AUCs of the machine learning features linear SVM, random forest, and gradient boost were 0.72, 0.78, and 0.82, respectively. Conclusion We found that a machine learning approach based on 18F-FDG textural features could predict the chemotherapy response using baseline PET images. This early prediction of the chemotherapy response may aid in determining treatment plans for osteosarcoma patients.

Incorporating crop phenological trajectory and texture for paddy rice detection with time series MODIS, HJ-1A and ALOS PALSAR imagery

ABSTRACT Repeated monitoring of paddy rice is essential for government agencies and policy makers to maintain the balance of supply and demand for rice. Recent studies have mostly concentrated on the mapping of paddy rice with temporal satellite imagery during growing seasons. Given the phenological variation within paddy rice fields and spectral confusion between paddies and other vegetation classes, our ability to identify paddy rice fields with temporal imagery remains limited. The objective of this study is to develop new phenology and textural-based strategies to detect paddies with HJ-1A, MODIS and PALSAR FNF imagery. Two phenology-based strategies that track the seasonal trajectory of crops and one textural-based strategy that contains image surface characteristics are presented. With the proposed strategies, temporal, spectral and textural features were investigated for paddy rice detection. The results indicate that the phenology-based strategies could reveal the phenological variation within paddy rice and significantly improved the detection accuracy. Seasonal amplitude, grey level co-occurrence matrix entropy and spectral features of the heading stage were proven to be important in identifying paddy rice. It was concluded that the combination of HJ-1A, MODIS and PALSAR FNF imagery are promising in facilitating the rapid mapping of paddy rice at a regional scale.

Investigating the histopathologic correlates of 18F-FDG PET heterogeneity in non-small-cell lung cancer

Despite the growing use of fluorine-18-fluorodeoxyglucose (F-FDG) PET texture analysis to measure intratumoural heterogeneity in cancer research, the biologic basis of F-FDG PET-derived texture variables is poorly understood. We aimed to assess correlations between F-FDG PET-derived texture variables and whole-slide image (WSI)-derived metrics of tumour cellularity and spatial heterogeneity. Twenty-two patients with non-small-cell lung cancer prospectively underwent F-FDG PET imaging before tumour resection. We tested nine F-FDG PET parameters: metabolically active tumour volume, total lesion glycolysis, mean standardized uptake value (SUVmean), first-order entropy, energy, skewness, kurtosis, grey-level co-occurrence matrix entropy and lacunarity (SUV-lacunarity). From the haematoxylin and eosin-stained WSIs, we derived mean tumour-cell density (MCD) and lacunarity (path-lacunarity). Spearman's correlation analysis and agglomerative hierarchical clustering were performed to assess variable associations. Tumour volumes ranged from 2.2 to 74 cm (median: 17.9 cm). MCD correlated positively with total lesion glycolysis (rs: 0.46, P: 0.007) and SUVmean (rs : 0.55; P: 0.008) and negatively with skewness and kurtosis (rs: -0.47 for both; P: 0.028 and 0.026, respectively). SUV-lacunarity and path-lacunarity were positively correlated (rs: 0.5; P: 0.018). On cluster analysis, larger tumours trended towards higher SUVmean and entropy with a predominance of tightly concentrated high SUV-voxels (negative skewness and low kurtosis on the histogram); on WSI analysis such larger tumours also displayed generally higher MCD and low SUV-lacunarity and path-lacunarity. Our data suggest that histopathological MCD and lacunarity are associated with several commonly used F-FDG PET-derived indices including SUV-lacunarity, metabolically active tumour volume, SUVmean, entropy, skewness, and kurtosis, and thus may explain the biological basis of F-FDG PET-uptake heterogeneity in non-small-cell lung cancer.

Role of Imaging in Predicting Radiation Induced Xerostomia

The development of predictive biomarkers for radiation toxicities (e.g. xerostomia) may better identify head and neck cancer (HNC) patients suitable for treatment intensification. Radiomics is a promising approach to identify image-based biomarkers. We hypothesize that baseline salivary gland structure-function evaluated using computed tomography (CT) and magnetic resonance imaging (MRI) based radiomics can be used for predicting radiation induced xerostomia. HNC patients treated between 2009 and 2015 at our institution with xerostomia graded by NCI-CTCAE at 3-month follow up were included. For each patient, contours of the parotid and submandibular glands on baseline planning CT (non-contrast enhanced) were prospectively collected. CT images were corrected for metal artifact reduction to minimize the effect of dental artifacts in radiomic feature computation. Baseline MR images were acquired post-contrast using a T1-weighted pulse sequence. CT images and corresponding contours were registered to MR images using rigid or deformable registration with commercially available software. In each patient, 5744 CT and MR features were extracted for ipsilateral(IL)/contralateral(CL) salivary glands. The dataset was randomly partitioned into training and validation sets using an 80/20 split. Features from the training set were first ranked in univariate analysis by likelihood for predicting severe xerostomia (grade ≥2). Features correlated with xerostomia (p<0.1) were further reduced using a maximum relevance minimum redundancy method. The final set of image features was selected using LASSO logistic regression. A Random Forest model with leave-one-out cross-validation was applied to predict radiation induced xerostomia. Three models were developed with image features extracted from (1) CT only, (2) MR only, and (3) CT+MR. Area under the receiver operator curve (AUC) was compared across models. The training dataset (n=112) revealed 11 CT and 6 MR image features. CT features included: CL parotid Gray Level Size Zone matrix (GLSZM) gray level non-uniformity (consistent with parotid gland heterogeneity) and Gray Level Co-occurrence Matrix (GLCM) correlation; IL submandibular GLCM entropy and total energy. MR features included: CL parotid Short Run High Gray Level Emphasis and total energy features. The cross-validated AUCs for the training data for CT only/MR only/CT+MR was 0.72/0.60/0.74 in xerostomia prediction. The AUC for the validation data (n=28) for CT only/MR only/CT+MR features was 0.82/0.68/0.73. In this cohort of HNC patients, these results suggest that CT image features may be used to predispose patients susceptible to radiation induced xerostomia. Further investigation with a larger cohort is required.

Prognostic value of radiomic analysis of iodine overlay maps from dual-energy computed tomography in patients with resectable lung cancer.

To investigate whether radiomics on iodine overlay maps from dual-energy computed tomography (DECT) can predict survival outcomes in patients with resectable lung cancer. Ninety-three lung cancer patients eligible for curative surgery were examined with DECT at the time of diagnosis. The median follow-up was 60.4 months. Radiomic features of the entire primary tumour were extracted from iodine overlay maps generated by DECT. A Cox proportional hazards regression model was used to determine independent predictors of overall survival (OS) and disease-free survival (DFS), respectively. Forty-two patients (45.2%) had disease recurrence and 39 patients (41.9%) died during the follow-up period. The mean DFS was 49.8 months and OS was 55.2 months. Univariate analysis revealed that significant predictors of both OS and DFS were stage and radiomic parameters, including histogram energy, histogram entropy, grey-level co-occurrence matrix (GLCM) angular second moment, GLCM entropy and homogeneity. The multivariate analysis identified stage and entropy as independent risk factors predicting both OS (stage, hazard ratio (HR) = 2.020 [95% CI 1.014-4.026], p = 0.046; entropy, HR = 1.543 [95% CI 1.069-2.228], p = 0.021) and DFS (stage, HR = 2.132 [95% CI 1.060-4.287], p = 0.034; entropy, HR = 1.497 [95% CI 1.031-2.173], p = 0.034). The C-index showed that adding entropy improved prediction of OS compared to stage only (0.720 and 0.667, respectively; p = 0.048). Radiomic features extracted from iodine overlay map reflecting heterogeneity of tumour perfusion can add prognostic information for patients with resectable lung cancer. • Radiomic feature (histogram entropy) from DECT iodine overlay maps was an independent risk factor predicting both overall survival and disease-free survival. • Adding histogram entropy to clinical stage improved prediction of overall survival compared to stage only (0.720 and 0.667, respectively; p = 0.048). • DECT can be a good option for comprehensive pre-operative evaluation in cases of resectable lung cancer.

Impact of Reconstruction Algorithms on CT Radiomic Features of Pulmonary Tumors: Analysis of Intra- and Inter-Reader Variability and Inter-Reconstruction Algorithm Variability

PurposeTo identify the impact of reconstruction algorithms on CT radiomic features of pulmonary tumors and to reveal and compare the intra- and inter-reader and inter-reconstruction algorithm variability of each feature.MethodsForty-two patients (M:F = 19:23; mean age, 60.43±10.56 years) with 42 pulmonary tumors (22.56±8.51mm) underwent contrast-enhanced CT scans, which were reconstructed with filtered back projection and commercial iterative reconstruction algorithm (level 3 and 5). Two readers independently segmented the whole tumor volume. Fifteen radiomic features were extracted and compared among reconstruction algorithms. Intra- and inter-reader variability and inter-reconstruction algorithm variability were calculated using coefficients of variation (CVs) and then compared.ResultsAmong the 15 features, 5 first-order tumor intensity features and 4 gray level co-occurrence matrix (GLCM)-based features showed significant differences (p<0.05) among reconstruction algorithms. As for the variability, effective diameter, sphericity, entropy, and GLCM entropy were the most robust features (CV≤5%). Inter-reader variability was larger than intra-reader or inter-reconstruction algorithm variability in 9 features. However, for entropy, homogeneity, and 4 GLCM-based features, inter-reconstruction algorithm variability was significantly greater than inter-reader variability (p<0.013).ConclusionsMost of the radiomic features were significantly affected by the reconstruction algorithms. Inter-reconstruction algorithm variability was greater than inter-reader variability for entropy, homogeneity, and GLCM-based features.

Temporal Changes of Texture Features Extracted From Pulmonary Nodules on Dynamic Contrast-Enhanced Chest Computed Tomography: How Influential Is the Scan Delay?

The aim of this study was to describe the temporal changes of various texture features extracted from pulmonary nodules on dynamic contrast-enhanced computed tomography (DCE-CT) and to compare the feature values among multiple scanning time points. We also aimed to analyze the variability of texture features across multiple scan delay times. This retrospective study was approved by the institutional review board of Seoul National University Hospital with waiver of patients' informed consent. Twenty patients (M:F, 6:14; mean age, 60.25 ± 11.97 years) with 20 lung nodules (mean size, 24.1 ± 12.3 mm) underwent DCE-CT with multiple scan delays (30, 60, 90, 120, 150, 180, 210, 240, 300, and 480 seconds) after precontrast scans. Lung nodule segmentation and texture feature extraction were performed at each time point using in-house software. Texture feature values were compared among the multiple time points using the Friedman test with post hoc pairwise Wilcoxon signed rank test. In addition, the dynamic range (DR) reflecting the variability between 2 time points to the interpatient range was calculated. Thereafter, we determined the stable time range that met both "DR greater than 0.90" and "no statistically significant difference" between all time point pairs for each feature. The degree of variability across all scan delay times was obtained using coefficients of variation. Standard deviation, variance, entropy, sphericity, discrete compactness, gray-level cooccurrence matrix (GLCM) inverse difference moment (IDM), GLCM contrast, and GLCM entropy did not show significant differences between scan delays of 30 and 180 seconds with DR greater than 0.90 between all time point pairs. When the range was narrowed down to 60 to 150 seconds, an additional 2 values (mean and homogeneity) showed stability. Among the 13 texture features, entropy, sphericity, discrete compactness, and GLCM entropy exhibited the lowest variability (coefficient of variation ≤5%). Most texture features exhibited stability with low variation between 60 and 150 seconds on DCE-CT. Thus, texture features extracted from contrast-enhanced CT with a scan delay range of 60 to 150 seconds can be used for tumor characterization despite the heterogeneity in delay time.

Gray level co-occurrence matrix algorithm as pattern recognition biosensor for oxidopamine-induced changes in lymphocyte chromatin architecture

We demonstrate that a proapoptotic chemical agent, oxidopamine, induces dose dependent changes in chromatin textural patterns which can be quantified using the Gray level co-occurrence matrix (GLCM) method. Peripheral blood (heparin-pretreated) samples were treated with oxidopamine (6-OHDA, 6-hydroxydopamine) to achieve effective concentrations of 100, 200 and 300µM. The samples were smeared on microscope slides and fixated in methanol. The smears were stained using a modification of Feulgen method for DNA visualization. For each stained smear, a sample of 30 lymphocyte chromatin structures were visualized and analyzed. This way, textural parameters for a total of 120 nuclei micrographs were calculated. For each chromatin structure, five different GLCM features were calculated: angular second moment, GLCM entropy, inverse difference moment, GLCM correlation, and GLCM variance. Oxidopamine induced the rise of the values of GLCM entropy and variance, and the reduction of angular second moment, correlation, and inverse difference moment. The trends for GLCM parameter changes were found to be highly significant (p<0.001). These results indicate that GLCM mathematical algorithm might be successfully used in detection and evaluation of discrete early apoptotic structural changes in Feulgen-stained chromatin of peripheral blood lymphocytes that are not detectable using conventional microscopy/cell biology techniques.

Relationship between chromatin complexity and nuclear envelope circularity in hippocampal pyramidal neurons

In this study we tested the existence and strength of the relationship between circularity of nuclear envelope and mathematical parameters of chromatin structure. Coronal sections of the brain were made in 10 male albino mice. The brain tissue was stained using a modification of Feulgen method for DNA visualization. A total of 200 hippocampal pyramidal neurons (20 per animal) were visualized using DEM 200 High-Speed Color CMOS Chip and Olympus CX21FS1 microscope. Circularity of the nuclear membrane was calculated in ImageJ (NIH, USA) after the nuclear segmentation, based on the freehand selection of the nuclear regions of interest. Circularity was determined from the values of area and perimeter. For each chromatin structure, using fractal and grey level co-occurrence matrix (GLCM) algorithms, we determined the values of fractal dimension, lacunarity, angular second moment, GLCM entropy, inverse difference moment, GLCM correlation, and GLCM contrast. It was found that circularity is in a significant correlation (p<0.05) with fractal dimension as the main parameter of fractal complexity analysis. Also, circularity was in a very strong relationship (p<0.001) with certain parameters of grey level co-occurrence matrix such as the angular second moment and GLCM correlation. This is the first study to indicate that nuclear shape is significantly related to mathematical parameters of higher chromatin organization. Also, it seems that circularity of the nuclear envelope is a good predictor of certain features of chromatin structure. Our study suggests that GLCM algorithm, as well as circularity analysis, have potentially substantial application in future cellular physiology research.

Çoklu ölçekli sayısal yükseklik modellerinden çıkarılan fizyografik detaylara ait yüzey doku özelliklerinin gri düzey eş-oluşum matrisi ile analizi

Analysis of surface textures of physiographic features extracted from multiscale digital elevation models via grey level co-occurrence matrix This paper is aimed at employing grey level co-occurrence matrix (GLCM) to analyse the surface textures of physiographic features extracted from multiscale digital elevation models (DEMs). Four GLCM parameters, energy, contrast, autocorrelation and entropy, are computed for horizontal (0°), vertical (90°) and diagonal (45 and 135°) cell pair orientations. For the respective DEMs and physiographic features, varying patterns are observed in the plots of the GLCM parameters due to varying surface profiles and the changes that occur over the scales. Due to the smoothing of the terrain during multiscaling, the features have increasing values of energy and entropy, and decreasing values of contrast and entropy, indicating decreasing roughness. Mountains have the highest roughness as compared to the other features over the scales, while basins have the lowest roughness. For each parameter, similar trends are observed in the plots for the four different cell pair orientations, indicating similar trends of change of surface texture in the different orientations over the scales. However, varying values are observed for the different orientations, depending on textural uniformity in the corresponding orientations. The results obtained demonstrate that GLCM can be an appropriate tool for classifying landforms from multiscale DEMs based on the different texture characteristics of the landforms.