Gray Level Co-occurrence Matrix Correlation Research Articles

Novel Functional Radiomics for Prediction of Cardiac PET Avidity in Lung Cancer Radiotherapy

Traditional methods of evaluating cardiotoxicity focus solely on radiation doses to the heart and do not incorporate functional imaging information. Functional imaging has great potential to improve the ability to provide early prediction for cardiotoxicity for lung cancer patients undergoing radiotherapy. FDG-based PET/CT imaging is routinely obtained as part of standard staging work up for lung cancer patients. Although FDG PET/CT scans are typically used to evaluate the tumor, imaging guidelines note that FDG PET/CT scans are an FDA-approved method to image for cardiac inflammation, and studies have noted that the PET cardiac signal can be predictive of clinical outcomes. The purpose of this work was to develop a radiomics model to predict clinical cardiac assessment of standard of care FDG PET/CT scans. The study included 100 consecutive lung cancer patients treated with radiotherapy who underwent standard pre-treatment FDG-PET/CT staging scans. A clinician reviewed the PET/CT scans per clinical cardiac assessment guidelines and classified the cardiac uptake as: 0 = uniform diffuse, 1 = absent, 2 = heterogeneous, with event rates of 20%, 44%, and 35%, respectively. The heart was delineated and 200 novel functional radiomics features were selected to classify cardiac FDG uptake patterns. We divided the data into an 80% training set and a 20% test set to train and evaluate the classification models. Feature reduction was carried out using the Wilcoxon test (with Bonferroni adjusted p<0.05), hierarchical clustering, and Recursive Feature Elimination. Two automatic machine learning (AutoML) frameworks were used to determine classification models: a Random Forest Classifier (Tree-based Pipeline Optimization Tool, TPOT) and Linear Discriminant Analysis (AutoSklearn). 10-fold cross validation was carried out for training and the accuracy of the ability of the models to predict for clinical cardiac assessment is reported. Fifty-one independent radiomics features were reduced to 3 clinically pertinent features (PET 2D Skewness, PET Grey Level Co-occurrence Matrix Correlation, and PET Median) using feature reduction techniques. The model selected by TPOT showed 89.8% predictive accuracy in the cross validation of the training set and 85% predictive accuracy on the test set. The model selected by AutoSklearn showed 89.7% predictive accuracy in the cross validation of the training set and 80% predictive accuracy on the test set. The novelty of this work is that it is the first study to develop and evaluate functional cardiac radiomic features from standard of care FDG PET/CT scans with the data showing good predictive accuracy with clinical imaging evaluation. If validated, the current work provides automated methods to provide functional cardiac information using standard of care imaging that can be used as an imaging biomarker for early clinical toxicity prediction for lung cancer patients.

Texture feature analysis of MRI-ADC images to differentiate glioma grades using machine learning techniques

Apparent diffusion coefficient (ADC) of magnetic resonance imaging (MRI) is an indispensable imaging technique in clinical neuroimaging that quantitatively assesses the diffusivity of water molecules within tissues using diffusion-weighted imaging (DWI). This study focuses on developing a robust machine learning (ML) model to predict the aggressiveness of gliomas according to World Health Organization (WHO) grading by analyzing patients’ demographics, higher-order moments, and grey level co-occurrence matrix (GLCM) texture features of ADC. A population of 722 labeled MRI-ADC brain image slices from 88 human subjects was selected, where gliomas are labeled as glioblastoma multiforme (WHO-IV), high-grade glioma (WHO-III), and low-grade glioma (WHO I-II). Images were acquired using 3T-MR systems and a region of interest (ROI) was delineated manually over tumor areas. Skewness, kurtosis, and statistical texture features of GLCM (mean, variance, energy, entropy, contrast, homogeneity, correlation, prominence, and shade) were calculated using ADC values within ROI. The ANOVA f-test was utilized to select the best features to train an ML model. The data set was split into training (70%) and testing (30%) sets. The train set was fed into several ML algorithms and selected most promising ML algorithm using K-fold cross-validation. The hyper-parameters of the selected algorithm were optimized using random grid search technique. Finally, the performance of the developed model was assessed by calculating accuracy, precision, recall, and F1 values reported for the test set. According to the ANOVA f-test, three attributes; patient gender (1.48), GLCM energy (9.48), and correlation (13.86) that performed minimum scores were excluded from the dataset. Among the tested algorithms, the random forest classifier(0.8772 ± 0.0237) performed the highest mean-cross-validation score and selected to build the ML model which was able to predict tumor categories with an accuracy of 88.14% over the test set. The study concludes that the developed ML model using the above features except for patient gender, GLCM energy, and correlation, has high prediction accuracy in glioma grading. Therefore, the outcomes of this study enable to development of advanced tumor classification applications that assist in the decision-making process in a real-time clinical environment.

Klasifikasi Citra Daging Babi dan Daging Kerbau Menggunakan Histogram Citra dan GLCM

Due to high consumer demand, some traders use the high price of meat to make a profit by mixing pork and buffalo meat. Some consumers are not aware of this, because in plain view buffalo meat with pork meat is difficult to distinguish, especially for some ordinary people. This action is very detrimental and disturbing the local community, especially Muslims. At present, technological advances in the field of digital image processing are increasing rapidly, especially in food products. In general, this research was conducted in 2 (three) stages. The first stage, namely the stage of image data collection of pork and buffalo meat. The second stage, namely the classification of pork and buffalo meat images using image histogram analysis and the Gray Level Co-occurrence Matrix (GLCM) method based on the color and texture of the meat. In this study using the Red Green Blue (RGB) color image method and GLCM texture extraction, namely contrast, homogeneity, energy, and correlation. The study was conducted using 20 samples of meat images (10 images of pork and 10 images of buffalo meat, respectively). Based on the results of the research that has been done, it was found that the image of buffalo meat has a higher percentage value of the Red (R) color component when compared to the pork image, whereas the percentage value of the Green (G) and Blue (B) color components is lower when compared to the image pork. Then, if the value between pixels is not homogeneous (small homogeneity value), then the contrast value is large, and vice versa if the value between pixels is homogeneous (large homogeneity value) then the contrast value is small. The image of buffalo meat has a small homogeneity value compared to the image of pork, so the variation in intensity (contrast) in the image of buffalo meat is high.

Can magnetic resonance imaging texture analysis change the breast imaging reporting and data system category of breast lesions?

PurposeThis study aimed to reveal magnetic resonance imaging (MRI) texture analysis (TA)'s contribution to categorizing breast lesions according to the Breast Imaging-Reporting and Data System (BI-RADS) lexicon. MethodTwo hundred and seventeen women with BI-RADS category 3, 4, and 5 lesions on breast MRI were included in the study. For TA, the region of interest was drawn manually to encompass the entire lesion on the fat-suppressed T2W and the first post-contrast T1W images. To identify the independent predictors of breast cancer, multivariate logistic regression analyses were performed using texture parameters. Estimated benign and malignant groups were formed according to the TA regression model. ResultsTexture parameters extracted from T2WI, including median, gray-level co-occurrence matrix (GLCM) contrast, GLCM correlation, GLCM joint entropy, GLCM sum entropy, and GLCM sum of squares, and parameters extracted from T1WI, including maximum, GLCM contrast, GLCM joint entropy, GLCM sum entropy, were independent predictors of breast cancer. In the estimated new groups according to the TA regression model, 19 (91%) of the benign 4a lesions were downgraded to BI-RADS category 3. ConclusionsThe addition of quantitative parameters obtained by MRI TA to BI-RADS criteria significantly increased the accuracy rate in differentiating benign and malignant breast lesions. When categorizing BI-RADS 4a lesions, the use of MRI TA in addition to conventional imaging findings may reduce unnecessary biopsy rates.

Texture analysis of the microstructure of internal curing concrete based on image recognition technology

Mechanical properties of internal curing concrete are greatly affected by its physical properties such as water content, cementing material content, porosity, and saturation. At the micro- level, such impact is finally reflected in the surface texture of its materials. In this study, the image recognition technology was used to find that the internal curing concrete samples have significant micromorphology and texture features. A texture parameter–strength model was established based on the relationship between Tamura texture parameters, gray level co-occurrence matrix (GLCM) texture parameters, and the mechanical strength. Due to the sensitivity of material properties and parameters, in terms of Tamura texture characteristics, coarseness, regularity and directionality are effective parameters for predicting the compressive strength of internally curing concrete. In terms of the GLCM texture, energy, correlation, entropy, and contrast are effective parameters to predict the compressive strength of the internal curing concrete. Correlations between each texture parameter and compressive strength follow different laws.

Discrimination between non-functioning pituitary adenomas and hypophysitis using machine learning methods based on magnetic resonance imaging‑derived texture features.

Hypophysitis is a heterogeneous condition that includes inflammation of the pituitary gland and infundibulum, and it can cause symptoms related to mass effects and hormonal deficiencies. We aimed to evaluate the potential role of machine learning methods in differentiating hypophysitis from non-functioning pituitary adenomas. The radiomic parameters obtained from T1A-C images were used. Among the radiomic parameters, parameters capable of distinguishing between hypophysitis and non-functioning pituitary adenomas were selected. In order to avoid the effects of confounding factors and to improve the performance of the classifiers, parameters with high correlation with each other were eliminated. Machine learning algorithms were performed with the combination of gray-level run-length matrix-low gray level run emphasis, gray-level co-occurrence matrix-correlation, and gray-level co-occurrence entropy. A total of 34 patients were included, 17 of whom had hypophysitis and 17 had non-functioning pituitary adenomas. Among the 38 radiomics parameters obtained from post-contrast T1-weighted images, 10 tissue features that could differentiate the lesions were selected. Machine learning algorithms were performed using three selected parameters; gray level run length matrix-low gray level run emphasis, gray-level co-occurrence matrix-correlation, and gray level co-occurrence entropy. Error matrices were calculated by using the machine learning algorithm and it was seen that support vector machines showed the best performance in distinguishing the two lesion types. Our analysis reported that support vector machines showed the best performance in distinguishing hypophysitis from non-functioning pituitary adenomas, emphasizing the importance of machine learning in differentiating the two lesions.

Gray-Level Co-occurrence Matrix Analysis for the Detection of Discrete, Ethanol-Induced, Structural Changes in Cell Nuclei: An Artificial Intelligence Approach.

Gray-level co-occurrence matrix (GLCM) analysis is a contemporary and innovative computational method for the assessment of textural patterns, applicable in almost any area of microscopy. The aim of our research was to perform the GLCM analysis of cell nuclei in Saccharomyces cerevisiae yeast cells after the induction of sublethal cell damage with ethyl alcohol, and to evaluate the performance of various machine learning (ML) models regarding their ability to separate damaged from intact cells. For each cell nucleus, five GLCM parameters were calculated: angular second moment, inverse difference moment, GLCM contrast, GLCM correlation, and textural variance. Based on the obtained GLCM data, we applied three ML approaches: neural network, random trees, and binomial logistic regression. Statistically significant differences in GLCM features were observed between treated and untreated cells. The multilayer perceptron neural network had the highest classification accuracy. The model also showed a relatively high level of sensitivity and specificity, as well as an excellent discriminatory power in the separation of treated from untreated cells. To the best of our knowledge, this is the first study to demonstrate that it is possible to create a relatively sensitive GLCM-based ML model for the detection of alcohol-induced damage in Saccharomyces cerevisiae cell nuclei.

The 3-Dimensional-Computed Tomography Texture Is Useful to Predict Pancreatic Neuroendocrine Tumor Grading.

The aim of this study is to evaluate the computed tomography texture parameters in predicting grading. This study analyzed 68 nonfunctioning pancreatic neuroendocrine neoplasms (Pan-NENs). Clinical and radiological parameters were studied. Four model models were built, including clinical and standard radiologic parameters (model 1), first- and second-order computed tomography features (models 2 and 3), all parameters (model 4). The diagnostic accuracy was reported as area under the curve. A score was computed using the best model and validated to predict progression-free survival. The size of tumors and heterogeneous enhancement were related to the risk of "non-G1" Pan-NENs (coefficients 0.471, P = 0.012, and 1.508, P = 0.027). Four second-order parameters were significantly related to the presence of "non-G1" Pan-NENs: the gray level co-occurrence matrix correlation (6.771; P = 0.011), gray level co-occurrence matrix contrast variance (0.349; P = 0.009), the neighborhood gray-level different matrix contrast (-63.129; P = 0.001), and the gray-level zone length matrix with the low gray-level zone emphasis (-0.151; P = 0.049). Model 4 was the best, with a higher area under the curve (0.912; P = 0.005). The score obtained predicted the progression-free survival. Computed tomography radiomics signature can be useful in preoperative workup.

Predicting Progression-Free Survival Using Primary and Nodal Radiomic Features in Head and Neck Cancer

Imaging quantified by radiomics has yielded reproducible predictions of both oncologic outcomes and treatment-related toxicity. Much of the correlation with oncologic outcomes has focused on image features from the primary tumor. However, both primary tumor and nodal features can impact outcomes, and it is known that differences in gene expression in primary tumors and nodal metastases provide distinct information regarding survival in head and neck squamous cell carcinomas (HNSCC). Nevertheless, current radiomic biomarkers often make no distinction between primary and nodal radiomic features. Therefore, we hypothesize that combining CT image features of primary tumors and nodal metastases in HNSCC will improve our ability to predict progression-free survival (PFS) compared to primary features alone.Pre-treatment CT images were prospectively collected from 115 HNSCC patients undergoing RT at our institution from 2008-2018. PFS was determined at time of last follow-up prior to 2018. Clinical data were systemically captured from patient medical charts. 107 radiomic features were extracted for the primary tumor and the ipsilateral/contralateral nodal regions. Features were pre-selected by literature review for those possibly significant in predicting loco-regional control of HNSCC. Tumor volume, correlation of gray level co-occurrence matrix (Corr-GLCM), run length non-uniformity of gray level of gray level run length matrix (RLN-GLRLM), and coarseness of neighboring gray tone difference matrix (Coarse-NGTDM) for the primary contours; and tumor volume and Corr-GLCM of the nodal contours were included in the model. A regression analysis of pre-selected features was performed using ridge regularization in the training set (n = 93 primary; n = 48 primary & nodal), and a generalized linear model was built with repeated ten-fold cross validation of patients treated from 2008-2017. Model performance was evaluated in a test set (patients treated 2017-2018) using the area under the receiver operating characteristic curve (AUC). Models were built separately for patients with primary tumor only and primary + nodal radiomic features.The AUC/sensitivity/specificity on the test set for the model with primary features (n = 22) was 0.91/1.00/0.75, and for the model with primary and nodal features (n = 18) was 0.75/0.60/0.92. The features with the highest coefficients in the combined model included primary Coarse-NGTDM (β = -0.21), primary Corr-GLCM (β = 0.0082), and nodal Corr-GLCM (β = -0.00041).In a small cohort of HNSCC patients, combining nodal features with primary features did not improve the prediction performance compared to primary features alone. This work suggests that primary tumor radiomic features may influence PFS more than nodal features. Future work is required in a larger cohort to validate these findings.

CT Radiomic Features of Superior Mesenteric Artery Involvement in Pancreatic Ductal Adenocarcinoma: A Pilot Study.

Background Current imaging methods for prediction of complete margin resection (R0) in patients with pancreatic ductal adenocarcinoma (PDAC) are not reliable. Purpose To investigate whether tumor-related and perivascular CT radiomic features improve preoperative assessment of arterial involvement in patients with surgically proven PDAC. Materials and Methods This retrospective study included consecutive patients with PDAC who underwent surgery after preoperative CT between 2012 and 2019. A three-dimensional segmentation of PDAC and perivascular tissue surrounding the superior mesenteric artery (SMA) was performed on preoperative CT images with radiomic features extracted to characterize morphology, intensity, texture, and task-based spatial information. The reference standard was the pathologic SMA margin status of the surgical sample: SMA involved (tumor cells ≤1 mm from margin) versus SMA not involved (tumor cells >1 mm from margin). The preoperative assessment of SMA involvement by a fellowship-trained radiologist in multidisciplinary consensus was the comparison. High reproducibility (intraclass correlation coefficient, 0.7) and the Kolmogorov-Smirnov test were used to select features included in the logistic regression model. Results A total of 194 patients (median age, 66 years; interquartile range, 60-71 years; age range, 36-85 years; 99 men) were evaluated. Aside from surgery, 148 patients underwent neoadjuvant therapy. A total of 141 patients' samples did not involve SMA, whereas 53 involved SMA. A total of 1695 CT radiomic features were extracted. The model with five features (maximum hugging angle, maximum diameter, logarithm robust mean absolute deviation, minimum distance, square gray level co-occurrence matrix correlation) showed a better performance compared with the radiologist assessment (model vs radiologist area under the curve, 0.71 [95% CI: 0.62, 0.79] vs 0.54 [95% CI: 0.50, 0.59]; P < .001). The model showed a sensitivity of 62% (33 of 53 patients) (95% CI: 51, 77) and a specificity of 77% (108 of 141 patients) (95% CI: 60, 84). Conclusion A model based on tumor-related and perivascular CT radiomic features improved the detection of superior mesenteric artery involvement in patients with pancreatic ductal adenocarcinoma. © RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Do and Kambadakone in this issue.

Usefulness of texture features of apparent diffusion coefficient maps in predicting chemoradiotherapy response in muscle-invasive bladder cancer.

To examine the usefulness of the texture analysis (TA) of apparent diffusion coefficient (ADC) maps in predicting the chemoradiotherapy (CRT) response of muscle-invasive bladder cancer (MIBC). We reviewed 45 MIBC patients who underwent cystectomy after CRT. CRT response was assessed through histologic evaluation of cystectomy specimens. Two radiologists determined the volume of interest for the index lesions on ADC maps of pretherapeutic 1.5-T MRI and performed TA using the LIFEx software. Forty-six texture features (TFs) were selected based on their contribution to the prediction of CRT sensitivity. To evaluate diagnostic performance, diagnostic models from the selected TFs were created using random forest (RF) and support vector machine (SVM), respectively. Twenty-three patients achieved pathologic complete response (pCR) to CRT. The feature selection identified first quartile ADC (Q1 ADC), gray-level co-occurrence matrix (GLCM) correlation, and GLCM homogeneity as important in predicting CRT response. Patients who achieved pCR showed significantly lower Q1 ADC and GLCM correlation values (0.66 × 10-3mm2/s and 0.53, respectively) than those who did not (0.81 × 10-3mm2/s and 0.70, respectively; p < 0.05 for both). The AUCs of the RF and SVM models incorporating the selected TFs were 0.82 (95% confidence interval [CI]: 0.67-0.97) and 0.96 (95% CI: 0.91-1.00), respectively, and the AUC of the SVM model was better than that of the mean ADC value (0.76, 95% CI: 0.61-0.90; p = 0.0037). TFs can serve as imaging biomarkers in MIBC patients for predicting CRT sensitivity. TAs of ADC maps can potentially optimize patient selection for CRT. • Texture analysis of ADC maps and feature selection identified important texture features for classifying pathologic tumor response in patients with muscle-invasive bladder cancer. • The machine learning model incorporating the texture features set, which included first quartile ADC, GLCM correlation, and GLCM homogeneity, showed high performance in predicting chemoradiotherapy response. • Texture features could serve as imaging biomarkers that optimize eligible patient selection for chemoradiotherapy in muscle-invasive bladder cancer.

Texture analysis for chemotherapy response evaluation in osteosarcoma using MR imaging.

The efficacy of MRI-based statistical texture analysis (TA) in predicting chemotherapy response among patients with osteosarcoma was assessed. Forty patients (male: female = 31:9; age = 17.2 ± 5.7 years) with biopsy-proven osteosarcoma were analyzed in this prospective study. Patients were scheduled for three cycles of neoadjuvant chemotherapy (NACT) and diffusion-weighted MRI acquisition at three time points: at baseline (t0), after the first NACT (t1) and after the third NACT (t2) using a 1.5 T scanner. Eight patients (nonsurvivors) died during NACT while 34 patients (survivors) completed the NACT regimen followed by surgery. Histopathological evaluation was performed in the resected tumor to assess NACT response (responder [≤50% viable tumor] and nonresponder [>50% viable tumor]) and revealed nonresponder: responder = 20:12. Apparent diffusion coefficient (ADC) and intravoxel incoherent motion (IVIM) parameters, diffusion coefficient (D), perfusion coefficient (D*) and perfusion fraction (f) were evaluated. A total of 25 textural features were evaluated on ADC, D, D* and f parametric maps and structural T1-weighted (T1W) and T2-weighted (T2W) images in the entire tumor volume using 3D TA methods gray-level cooccurrence matrix (GLCM), neighborhood gray-tone-difference matrix (NGTDM) and run-length matrix (RLM). Receiver-operating-characteristic curve analysis was performed on the selected textural feature set to assess the role of TA features (a) as marker(s) of tumor aggressiveness leading to mortality at baseline and (b) in predicting the NACT response among survivors in the course of treatment. Findings showed that the NGTDM features coarseness, busyness and strength quantifying tumor heterogeneity in D, D* and f maps and T1W and T2W images were useful markers of tumor aggressiveness in identifying the nonsurvivor group (area-under-the-curve [AUC] = 0.82-0.88) at baseline. The GLCM features contrast and correlation, NGTDM features contrast and complexity and RLM feature short-run-low-gray-level-emphasis quantifying homogeneity/terogeneity in tumor were effective markers for predicting chemotherapeutic response using D (AUC = 0.80), D* (AUC = 0.80) and T2W (AUC = 0.70) at t0, and D* (AUC = 0.80) and f (AUC = 0.70) at t1. 3D statistical TA features might be useful as imaging-based markers for characterizing tumor aggressiveness and predicting chemotherapeutic response in patients with osteosarcoma.

Parcel-level mapping of crops in a smallholder agricultural area: A case of central China using single-temporal VHSR imagery

Timely and accurate monitoring of crop patterns in smallholder agricultural areas is essential for guiding local crop yield estimates, agricultural subsidy allocations, and food security policy formulation. The household-operated land management model leads to a highly fragmented and heterogeneous agricultural landscape; therefore, fine crop mapping in smallholder agricultural areas remains challenging. By using very high spatial resolution (VHSR) images, this study aimed to explore parcel-level crop mapping methods using the case of a typical smallholder agricultural area in Wuhan, China. Object-based image analysis techniques as well as random forest (RF) and support vector machine (SVM) classifiers were used to classify WorldView-2 (WV2) images into eight crop-level agricultural land use categories. Several classification models were built using the combination of two classifiers and different image features, including spectral, geometrical, and textural features. The results showed that the classification model using the RF classifier and all 27 selected features had the highest accuracy, with an overall accuracy of 80.04% and a kappa value of 0.78; specifically, the user’s and producer’s accuracies of rice, cotton, lotus, bare paddy field and bare upland field exceeded 80%. We found that the performance of the RF and SVM classifiers was generally comparable, although as the input features increased, the accuracy of the RF was slightly higher than that of the SVM. The use of spatial features, such as the gray level cooccurrence matrix (GLCM) standard deviation, GLCM correlation, and area of image objects, could help improve the accuracy of parcel-level crop mapping. Our research confirmed the practical value of single-temporal VHRS images and RF classifiers in mapping parcel-level crops in complex agricultural areas. This framework provides a methodological reference for accurately monitoring crop distribution in smallholder agriculture areas to support the development of local precision agriculture.

Association Between the Size and 3D CT-Based Radiomic Features of Breast Cancer Hepatic Metastasis

To evaluate the effect of the anatomic size on 3D radiomic imaging features of the breast cancer hepatic metastases. CT scans of 81 liver metastases from 54 patients with breast cancer were evaluated. Ten most common 3D radiomic features from the histogram and gray level co-occurrence matrix (GLCM) categories were calculated for the hepatic metastases (HM) and compared to normal liver (NL). The effect of size was evaluated by using linear mixed-effects regression models. The effect of size on different radiomic features was analyzed for both liver lesions and background liver. Three-dimensional radiomic features from GLCM demonstrate an important size dependence. The texture-feature size dependence was found to be different among feature categories and between the HM and NL, thus demonstrating a discriminatory power for the tissue type. Significant difference in the slope was found for GLCM homogeneity (NL slope = 0.004, slope difference 95% confidence interval [CI] 0.06-0.1, p <0.001), contrast (NL slope = 45, slope difference 95% CI 205-305, p <0.001), correlation (NL slope = 0.04, slope difference 95% CI 0.11-0.21, p <0.001), and dissimilarity (NL slope = 0.7, slope difference 95% CI 3.6-5.4, p <0.001). The GLCM energy (NL slope = 0.002, slope difference 95% CI -0.0005 to -0.0003, p <0.007), and entropy (NL slope = 1.49, slope difference 95% CI 0.07-0.52, p <0.009) exhibited size-dependence for both NL and HM, although demonstrating a difference in the slope between themselves. Radiomic features of breast cancer hepatic metastasis exhibited significant correlation with tumor size. This finding demonstrates the complex behavior of imaging features and the need to include feature-specific properties into radiomic models.

Radiomics of cholangiocarcinoma on pretreatment CT can identify patients who would best respond to radioembolisation.

Results after trans-arterial radioembolisation (TARE) for intrahepatic cholangiocarcinoma (iCC) depend on the architecture of the tumour. This latter can be quantified through computed tomography (CT) texture analysis. The aims of the present study were to analyse relationships between CT textural features prior to TARE and objective response (OR), progression-free survival (PFS), and overall survival (OS). Texture analysis was retrospectively applied to 55 pre-TARE CT scans of iCCs, focusing attention on the histogram-based features and the grey-level co-occurrence matrix (GLCM). Texture features were harmonised using the ComBat procedure. Objective response was assessed using the Response Evaluation Criteria In Solid Tumours 1.1. The least absolute shrinkage and selection operator (LASSO) method was applied to select the most useful textural features related to OR. Of the 55 patients, 53 had post-TARE imaging available, showing OR in 56.6% of cases. Texture analysis showed that iCCs showing OR after TARE had a higher uptake of iodine contrast in the arterial phase (higher mean histogram values, p< 0.001) and more homogeneous distribution (lower kurtosis, p= 0.043; GLCM contrast, p= 0.004; GLCM dissimilarity, p= 0.005, and higher GLCM homogeneity, p= 0.005; and GLCM correlation p= 0.030) at the pre-TARE CT scan. A favourable radiomic signature was calculated and observed in 15 of the 55 patients. The median PFS of these 15 patients was 12.1months and that of the remaining 40 patients was 5.1months (p= 0.008). Texture analysis of pre-TARE CT scans can quantify vascularisation and homogeneity of iCC architecture, providing clinical information useful in identifying ideal TARE candidates. • Hypervascular tumours with a more homogeneous uptake of iodine contrast in the arterial phase were those most likely to be effectively treated by TARE. • The arterial phase was observed to be the best acquisition phase for providing information regarding the "sensitivity" of the tumour to TARE. • Patients with favourable radiomic signature showed a median progression-free survival of 12.1months versus 5.1months of patients with an unfavourable signature (p = 0.008).

Hypovascular pancreas head adenocarcinoma: CT texture analysis for assessment of resection margin status and high-risk features.

To determine if CT texture analysis features are associated with hypovascular pancreas head adenocarcinoma (PHA) postoperative margin status, nodal status, grade, lymphovascular invasion (LVI), and perineural invasion (PNI). This Research Ethics Board-approved retrospective cohort study included 131 consecutive patients with resected PHA. Tumors were segmented on preoperative contrast-enhanced CT. Tumor diameter and texture analysis features including mean, minimum and maximum Hounsfield units, standard deviation, skewness, kurtosis, and entropy and gray-level co-occurrence matrix (GLCM) features correlation and dissimilarity were extracted. Two-sample t test and logistic regression were used to compare parameters for prediction of margin status, nodal status, grade, LVI, and PNI. Diagnostic accuracy was assessed using receiver operating characteristic curves and Youden method was used to establish cutpoints. Margin status was associated with GLCM correlation (p = 0.012) and dissimilarity (p = 0.003); nodal status was associated with standard deviation (p = 0.026) and entropy (p = 0.031); grade was associated with kurtosis (p = 0.031); LVI was associated with standard deviation (p = 0.047), entropy (p = 0.026), and GLCM correlation (p = 0.033) and dissimilarity (p = 0.011). No associations were found for PNI (p > 0.05). Logistic regression yielded an area under the curve of 0.70 for nodal disease, 0.70 for LVI, 0.68 for grade, and 0.65 for margin status. Optimal sensitivity/specificity was as follows: nodal disease 73%/72%, LVI 72%/65%, grade 55%/83%, and margin status 63%/66%. CT texture analysis features demonstrate fair diagnostic accuracy for assessment of hypovascular PHA nodal disease, LVI, grade, and postoperative margin status. Additional research is rapidly needed to identify these high-risk features with better accuracy. • CT texture analysis features are associated with pancreas head adenocarcinoma postoperative margin status which may help inform treatment decisions as a negative resection margin is required for cure. • CT texture analysis features are associated with pancreas head adenocarcinoma nodal disease, a poor prognostic feature. • Indicators of more aggressive pancreas head adenocarcinoma biology including tumor grade and LVI can be diagnosed using CT texture analysis with fair accuracy.

Relationship of CT Radiomics and Dose Texture to Radiation-Induced Swallow Dysfunction in Head and Neck Cancer

Recent studies have shown that radiotherapy (RT) dosimetry is associated with an increased incidence of dysphagia in head and neck cancer (HNC) patients. The purpose of this work was to develop a dysphagia model using a comprehensive atlas of swallow-related structures to determine the predictive value of imaging and dose features of HNC patients who were evaluated for dysphagia after RT. We hypothesized that imaging and dosimetry features would be significantly correlated to patient reported scores of dysphagia. CT images and dysphagia scores were prospectively collected in HNC patients. We used the 3 month post-RT patient reported Sydney Swallow Questionnaire (SSQ) that has been validated to measure pharyngeal swallow as the primary endpoint. CT images were acquired prior to RT at the time of simulation. Swallow-related structures identified as important to swallow function (e.g. ipsi/contralateral parotid and submandibular glands [iPG, cPG, iSG, cSG], middle constrictors [MPC], anterior and posterior digastrics [ADM, PDM], mylohyoid [MHM], genioglossus [GGM], hyoglossus [HM], and cricopharyngeus [CP]) were segmented. For each segmented salivary gland, high-dimensional image features were extracted from CT images using a python software package. For each swallow-related structure, high-dimensional features were extracted from the structure’s dose distribution. Major categories of extracted features included shape, first order statistics, gray level co-occurrence matrix (GLCM), gray level run length matrix (GLRLM), and gray level size zone matrix (GLSZM). Multivariate regression was performed using the Enter method to identify the predictors of SSQ. The variables considered for modeling were chosen based on statistically significant univariate relationships with SSQ (p<0.005). All statistical analysis was performed in statistical software. In 136 patients, total SSQ at 3 months post-RT was correlated with the MHM D10, GLCM informational measure of correlation of dose, and GLSZM gray level non-uniformity of dose; cSG minimum dose and D10, HM minimum dose; and MPC GLCM inverse difference of dose (r>0.30, p<0.005). In terms of CT image features, SSQ was related to cPG GLCM cluster shade and prominence, iSG GLCM difference variance, and iPG GLCM correlation (r>0.30, p<0.005). In a multivariate analysis (R=0.65, p<0.001), CT iPG GLCM correlation, iSG GLCM difference variance, MPC inverse difference of dose and MHM D10 contributed significantly to the model (p<0.05). Salivary gland image features of CT intensity heterogeneity such as the iSG difference variance as well as the dose to 10% of the MHM and homogeneity of the dose to the MPC (identified by the inverse difference of dose) were significant predictors of dysphagia post-RT. These results suggest that both salivary structure and dose to swallow muscles are important dysphagia predictors and the promise of this approach for a comprehensive RT-dysphagia model.

Rock Characterization Using Gray‐Level Co‐Occurrence Matrix: An Objective Perspective of Digital Rock Statistics

AbstractModeling flow and transport in porous media using pore‐scale modeling is reliant on rock properties derived from digital rock images using segmentation techniques. These digital rock images obtained using computed tomography incorporate the variation in the intensity of phases depending on the attenuation of X‐rays. A standard technique is the segmentation of tomographic images based on user‐selected grayscale thresholding, allowing the identification of different phases. This threshold is subjective based on the operator and results in loss of essential information about the grayscale variation after segmentation. This paper implements the gray‐level co‐occurrence matrix (GLCM) incorporating the full range of grayscale information. The GLCM captures the relative occurrence of grayscale values in a spatial map. These maps show visually connected/disconnected populations of different phases such as pore space, quartz grains, minerals, and other features. We show that each rock has its own GLCM signature depending on the variations in gray‐level intensities. Several statistical measures are calculated: (1) GLCM contrast describing local variation in the gray‐level intensities, (2) GLCM angular second moment, describing the rock homogeneity; (3) GLCM mean, describing weighted average of the probability of occurrence of features based on their location on the GLCM map; and (4) GLCM correlation, measuring the linear dependencies of grayscale values and the degree of (an) isotropy in the micro–computed tomographic images of each of the rock types. The GLCM method provides a pathway to alleviate user biases and allow automation of micro–computed tomography analyses.

Feature Analysis of Kidney Ultrasound Image in Four Different Ultrasound using Gray Level Co-occurrence Matrix (GLCM) and Intensity Histogram (IH)

Misinterpretation analysis of ultrasound images has been huge issues in the world nowadays. Lack of skills and knowledge, as well as unclear ultrasound image due to the presence of speckle noise in ultrasound, are some factors lead to this issue. In this research, we compare 188 kidney ultrasound images from four different types of ultrasound machines, named as ultrasound A, B, C and D. Image pre-processing of images which involve cropping, enhancement, and filtering are performed before manual segmentation and texture analysis process to indicates the wanted region and improve contrast in each image. Texture analysis is performed using gray level co-occurrence matrix (GLCM) and intensity histogram (IH) to find differences and similarities in kidney image texture between all ultrasounds. Four GLCM parameters, contrast, correlation, energy and homogeneity and four parameters from IH (mean, standard deviation, variance, and skewness) used to indicate the most significant features between all ultrasound machines. Results show that contrast in GLCM is the most significant features that can be extracted from all four ultrasound machines and will be used in the classification process.

Automatic seam pucker evaluation using support vector machine classifiers

PurposeThe purpose of this paper is to propose a novel method using support vector machine (SVM) classifiers for objective seam pucker evaluation. Features are extracted using wavelet analysis and gray-level co-occurrence matrix (GLCM), and the samples are evaluated using SVM classifiers. The study aims to solve the problem of inappropriate parameters and large required samples in objective seam pucker evaluation.Design/methodology/approachInitially, seam pucker image was captured, and Edge detection and Hough transform were utilized to normalize the seam position and orientation. After cropping the image, the intensity was adjusted to the same identical level through histogram specification. Then, the standard deviations of the horizontal image and diagonal image, reconstructed using wavelet decomposition and reconstruction, were calculated based on parameter optimization. Meanwhile, GLCM was extracted from the restructured horizontal detail image, then the contrast and correlation of GLCM were calculated. Finally, these four features were imported to SVM classifiers based on genetic algorithm for evaluation.FindingsThe four extracted features reflected linear relationships among five grades. The experimental results showed that the classification accuracy was 96 percent, which catches up to the performance of human vision, and resolves ambiguity and subjective of the manual evaluation.Originality/valueThere are large required samples in current research. This paper provides a novel method using finite samples, and the parameters of the methods were discussed for parameter optimization. The evaluation results can provide references for analyzing the reason of wrinkles during garment manufacturing.