Use Of Texture Analysis Research Articles

Texture Analysis in Brain Tumor MR Imaging.

Texture analysis, as well as its broader category radiomics, describes a variety of techniques for image analysis that quantify the variation in surface intensity or patterns, including some that are imperceptible to the human visual system. Cerebral gliomas have been most rigorously studied in brain tumors using MR-based texture analysis (MRTA) to determine the correlation of various clinical measures with MRTA features. Promising results in cerebral gliomas have been shown in the previous MRTA studies in terms of the correlation with the World Health Organization grades, risk stratification in gliomas, and the differentiation of gliomas from other brain tumors. Multiple MRTA studies in gliomas have repeatedly shown high performance of entropy, a measure of the randomness in image intensity values, of either histogram- or gray-level co-occurrence matrix parameters. Similarly, researchers have applied MRTA to other brain tumors, including meningiomas and pediatric posterior fossa tumors.However, the value of MRTA in the clinical use remains undetermined, probably because previous studies have shown only limited reproducibility of the result in the real world. The low-to-modest generalizability may be attributed to variations in MRTA methods, sampling bias that originates from single-institution studies, and overfitting problems to a limited number of samples.To enhance the reliability and reproducibility of MRTA studies, researchers have realized the importance of standardizing methods in the field of radiomics. Another advancement is the recent development of a comprehensive assessment system to ensure the quality of a radiomics study. These two-way approaches will secure the validity of upcoming MRTA studies. The clinical use of texture analysis in brain MRI will be accelerated by these continuous efforts.

Usefulness of texture analysis in 123I-FP-CIT for the differentiation of AD and DLB subtypes.

I-2β-carbomethoxy-3β-(4-iodophenyl)-N-(3-fluoropropyl) nortropane (123I-FP-CIT) is well known to be a useful tracer for differentiating dementia with Lewy bodies (DLB) and Alzheimer disease (AD). However, clinically, there are some cases in which these diseases cannot be differentiated by ordinary quantitative methods. Therefore, in this study, we established an index that reflects not only the total count but also the distribution and heterogeneity of tracer uptake. We investigated whether assessment of the heterogeneous depletion of 123I-FP-CIT is useful for the differentiation of various types of dementia, i.e., probable DLB, possible DLB, and AD, using texture analysis. A total of 122 patients with either probable DLB (n=35), possible DLB (n=23), AD (n=44), and normal controls (n=20) were analyzed. Summated single photon emission computed tomography (SPECT) images (7 to 10 slices) of the patients, including the bilateral striatum, were analyzed using the gray-level histogram method (GLHM) of texture analysis. Mean, variance, skewness, and kurtosis of GLHM were compared with the specific binding ratio by Livia Tossici-Bolt's method (SBR). The sensitivity and specificity for differentiating probable DLB from possible DLB, AD, and normal controls were 97.1% and 77.0%, respectively, for skewness, using a cut-off point of 6.8%, and 97.1% and 81.6%, respectively, for kurtosis, using a cut-off point of 53.4%. The sensitivity and specificity for differentiating probable and possible DLB from AD and normal controls was 65.5% and 98.4%, respectively, for skewness, using a cut-off point of 6.4%, and 79.3% and 93.8%, respectively, for kurtosis, using a cut-off point of 53.4%. In the assessment of the efficacy of 123I-FP-CIT to differentiate AD and DLB subtypes, mean, variance, skewness, and kurtosis by GLHM was as useful as the SBR method. Moreover, possible DLB and probable DLB could be differentiated by skewness and kurtosis. Our results demonstrate that texture analysis is more useful than conventional quantitative methods for obtaining valuable information of the brain. Textural features as such may have considerable potential as imaging biomarkers of DLB progression.

Texture analysis of iodine maps and conventional images for k-nearest neighbor classification of benign and metastatic lung nodules

BackgroundThe purpose of this study was to analyze if the use of texture analysis on spectral detector CT (SDCT)-derived iodine maps (IM) in addition to conventional images (CI) improves lung nodule differentiation, when being applied to a k-nearest neighbor (KNN) classifier.Methods183 cancer patients who underwent contrast-enhanced, venous phase SDCT of the chest were included: 85 patients with 146 benign lung nodules (BLN) confirmed by either prior/follow-up CT or histopathology and 98 patients with 425 lung metastases (LM) verified by histopathology, 18F-FDG-PET-CT or unequivocal change during treatment. Semi-automatic 3D segmentation of BLN/LM was performed, and volumetric HU attenuation and iodine concentration were acquired. For conventional images and iodine maps, average, standard deviation, entropy, kurtosis, mean of the positive pixels (MPP), skewness, uniformity and uniformity of the positive pixels (UPP) within the volumes of interests were calculated. All acquired parameters were transferred to a KNN classifier.ResultsDifferentiation between BLN and LM was most accurate, when using all CI-derived features combined with the most significant IM-derived feature, entropy (Accuracy:0.87; F1/Dice:0.92). However, differentiation accuracy based on the 4 most powerful CI-derived features performed only slightly inferior (Accuracy:0.84; F1/Dice:0.89, p=0.125). Mono-parametric lung nodule differentiation based on either feature alone (i.e. attenuation or iodine concentration) was poor (AUC=0.65, 0.58, respectively).ConclusionsFirst-order texture feature analysis of contrast-enhanced staging SDCT scans of the chest yield accurate differentiation between benign and metastatic lung nodules. In our study cohort, the most powerful iodine map-derived feature slightly, yet insignificantly increased classification accuracy compared to classification based on conventional image features only.

Review of Value of CT Texture Analysis and Machine Learning in Differentiating Fat-Poor Renal Angiomyolipoma from Renal Cell Carcinoma.

The diagnosis of patients with suspected angiomyolipoma relies on the detection of abundant macroscopic intralesional fat, which is always of no use to differentiate fat-poor angiomyolipoma (fp-AML) from renal cell carcinoma and diagnosis of fp-AML excessively depends on individual experience. Texture analysis was proven to be a potentially useful biomarker for distinguishing between benign and malignant tumors because of its capability of providing objective and quantitative assessment of lesions by analyzing features that are not visible to the human eye. This review aimed to summarize the literature on the use of texture analysis to diagnose patients with fat-poor angiomyolipoma vs those with renal cell carcinoma and to evaluate its current application, limitations, and future challenges in order to avoid unnecessary surgical resection.

Differentiation of periapical granuloma from radicular cyst using cone beam computed tomography images texture analysis

ObjectiveThis study aimed to investigate the use of texture analysis for characterization of radicular cysts and periapical granulomas and to assess its efficacy to differentiate between both lesions with histological diagnosis. MethodsCone beam computed tomography (CBCT) images were obtained from 19 patients with 25 periapical lesions (14 radicular cysts and 11 periapical granulomas) confirmed by biopsy. Regions of interest were created in the lesions from which 11 texture parameters were calculated. Spearman's correlation analysis was performed and adjusted with Benjamini-Hochberg false discovery rate procedure (FDR <0.005). ResultsThe texture parameters used to differentiate the lesions were assessed by using a receiver operating characteristic analysis. Five texture parameters were predictive of lesion differentiation for eight positions: angular second moment; sum of squares; sum of average; contrast; correlation. ConclusionTexture analysis of CBCT scans distinguishes radicular cysts from periapical granulomas and can be a promising diagnostic tool for periapical lesions. Clinical significanceTexture analysis can be used in diagnostic and treatment monitoring to provide supplementary information.

Usefulness of texture analysis for grading pancreatic neuroendocrine tumors on contrast-enhanced computed tomography and apparent diffusion coefficient maps

To determine whether texture analysis of contrast-enhanced computed tomography (CECT) and apparent diffusion coefficient (ADC) maps could predict tumor grade (G1 vs G2-3) in patients with pancreatic neuroendocrine tumor (PNET). Thirty-three PNETs (22 G1 and 11 G2-3) were retrospectively reviewed. Fifty features were individually extracted from the arterial and portal venous phases of CECT and ADC maps by two radiologists. Diagnostic performance was assessed by receiver operating characteristic curves while inter-observer agreement was determined by calculating intraclass correlation coefficients (ICCs). G2-G3 tumors were significantly larger than G1. Seventeen features significantly differed among the two readers on univariate analysis, with ICCs > 0.6; the largest area under the curve (AUC) for features of each CECT phase and ADC map was log-sigma 1.0 joint-energy = 0.855 for the arterial phase, log-sigma 1.5 kurtosis = 0.860 for the portal venous phase, and log-sigma 1.0 correlation = 0.847 for the ADC map. The log-sigma 1.5 kurtosis of the portal venous phase showed the largest AUC in the CECT and ADC map, and its sensitivity, specificity, and accuracy were 95.5%, 72.7%, and 87.9%, respectively. Texture analysis may aid in differentiating between G1 and G2-3 PNET.

Texture Analysis as a Radiomic Marker for Differentiating Benign From Malignant Adrenal Tumors.

The aim of this study was to evaluate the use of texture analysis for differentiation between benign from malignant adrenal lesions on contrast-enhanced abdominal computed tomography (CT). After institutional review board approval, a retrospective analysis was performed, including an electronic search of pathology records for all biopsied adrenal lesions. Patients were included if they also had a contrast-enhanced abdominal CT in the portal venous phase. Computed tomographic images were manually segmented, and texture analysis of the segmented tumors was performed. Texture analysis results of benign and malignant tumors were compared, and areas under the curve (AUCs) were calculated. One hundred twenty-five patients were included in the analysis. Excellent discriminators of benign from malignant lesions were identified, including entropy and standard deviation. These texture features demonstrated lower values for benign lesions compared with malignant lesions. Entropy values of benign lesions averaged 3.95 using a spatial scaling factor of 4 compared with an average of 5.08 for malignant lesions (P < .0001). Standard deviation values of benign lesions averaged 19.94 on the unfiltered image compared with an average of 34.32 for malignant lesions (P < .0001). Entropy demonstrated AUCs ranging from 0.95 to 0.97 for discriminating tumors, with sensitivities and specificities ranging from 81% to 95% and 88% to 100%, respectively. Standard deviation demonstrated AUCs ranging from 0.91 to 0.94 for discriminating tumors, with sensitivities and specificities ranging from 73% to 93% and 86% to 95%, respectively. Texture analysis offers a noninvasive tool for differentiating benign from malignant adrenal tumors on contrast-enhanced CT images. These results support the further development of texture analysis as a quantitative biomarker for characterizing adrenal tumors.

Decoding incidental ovarian lesions: use of texture analysis and machine learning for characterization and detection of malignancy.

To compare CT texture features of benign and malignant ovarian lesions and to build a machine learning model to detect malignancy in incidental ovarian lesions. In this IRB-approved, HIPAA-compliant, retrospective study, 427 consecutive patients with incidental ovarian lesions detected on contrast-enhanced CT (348, 81.5% benign and 79, 18.5% malignant) were included. The following CT texture features were analyzed using commercially available software (TexRAD, Feedback Plc, Cambridge, UK): total pixel, mean, standard deviation (SD), entropy, mean value of positive pixels (MPP), skewness, kurtosis and entropy. Three machine learning models were created by combining texture features and patients' age, and performance of these models was assessed using tenfold cross-validation. Receiver operating characteristics (ROC) were constructed to assess sensitivity and specificity. The cutoff value was picked using a cost-weighted method. Total pixels, mean, SD, entropy, MPP, and skewness were significantly different between benign and malignant groups (p < 0.05). With a selected 10 as a cost factor to optimize cutoff value selection, sensitivity 92%, specificity 60% in the random forest (RF) model, sensitivity 91%, specificity 69% in SVM model, and sensitivity 92%, specificity 61% in the logistic regression, respectively. CT texture analysis could provide objective imaging analysis of incidental ovarian lesions and ML models using CT texture features and age demonstrated high sensitivity and moderate specificity for detection of malignant lesions.

Liver Tumor Burden in Pancreatic Neuroendocrine Tumors: CT Features and Texture Analysis in the Prediction of Tumor Grade and 18F-FDG Uptake.

Pancreatic neuroendocrine tumors (p-NETs) are a rare group of neoplasms that often present with liver metastases. Histological characteristics, metabolic behavior, and liver tumor burden (LTB) are important prognostic factors. In this study, the usefulness of texture analysis of liver metastases in evaluating the biological aggressiveness of p-NETs was assessed. Fifty-six patients with liver metastases from p-NET were retrospectively enrolled. Qualitative and quantitative CT features of LTB were evaluated. Histogram-derived parameters of liver metastases were calculated and correlated with the tumor grade (G) and 18F-fluorodeoxyglucose (18F-FDG) standardized uptake value (SUV). Arterial relative enhancement was inversely related with G (−0.37, p = 0.006). Different metastatic spread patterns of LTB were not associated with histological grade. Arterialentropy was significantly correlated to G (−0.368, p = 0.038) and to Ki67 percentage (−0.421, p = 0.018). The ROC curve for the Arterialentropy reported an area under the curve (AUC) of 0.736 (95% confidence interval 0.545–0.928, p = 0.035) in the identification of G1–2 tumors. Arterialuniformity values were correlated to G (0.346, p = 0.005) and Ki67 levels (0.383, p = 0.033). Arterialentropy values were directly correlated with the SUV (0.449, p = 0.047) which was inversely correlated with Arterialuniformity (−0.499, p = 0.025). Skewness and kurtosis reported no significant correlations. In conclusion, histogram-derived parameters may predict adverse histological features and metabolic behavior of p-NET liver metastases.

Detection of Myocardial Tissue Alterations in Hypertrophic Cardiomyopathy Using Texture Analysis of T2-Weighted Short Inversion Time Inversion Recovery Magnetic Resonance Imaging.

The aim of this study was to evaluate the usefulness of texture analysis of T2-weighted short inversion time inversion recovery (T2-STIR) for detecting myocardial tissue alterations in hypertrophic cardiomyopathy (HCM). Twenty patients with HCM and 11 controls were examined. Texture analysis was performed for the hypertrophied regions with and without and abnormal hyperintensity (AHI) and for the interventricular septum of the controls on T2-STIR. T2 mapping was performed to measure myocardial T2 values. A gray-level nonuniformity value of 64.7 was the best discriminator between patients and controls with an area under the curve of 0.93 on a receiver operating characteristic curve. T2 values did not differ between them. The gray-level nonuniformity was significantly smaller in AHI regions than in the hypertrophied regions without AHI in HCM patients. Texture analysis is useful for quantitatively detecting myocardial tissue altenations, including AHI, associated with HCM on T2-STIR.

CT Texture Analysis Challenges: Influence of Acquisition and Reconstruction Parameters: A Comprehensive Review.

Texture analysis in medical imaging is a promising tool that is designed to improve the characterization of abnormal images from patients, to ultimately serve as a predictive or prognostic biomarker. However, the nature of image acquisition itself implies variability in each pixel/voxel value that could jeopardize the usefulness of texture analysis in the medical field. In this review, a search was performed to identify current published data for computed tomography (CT) texture reproducibility and variability. On the basis of this analysis, the critical steps were identified with a view of using texture analysis as a reliable tool in medical imaging. The need to specify the CT scanners used and the associated parameters in published studies is highlighted. Harmonizing acquisition parameters between studies is a crucial step for future texture analysis.

Differentiating Grade in Breast Invasive Ductal Carcinoma Using Texture Analysis of MRI.

Purpose The objective of this study is to investigate the use of texture analysis (TA) of magnetic resonance image (MRI) enhanced scan and machine learning methods for distinguishing different grades in breast invasive ductal carcinoma (IDC). Preoperative prediction of the grade of IDC can provide reference for different clinical treatments, so it has important practice values in clinic. Methods Firstly, a breast cancer segmentation model based on discrete wavelet transform (DWT) and K-means algorithm is proposed. Secondly, TA was performed and the Gabor wavelet analysis is used to extract the texture feature of an MRI tumor. Then, according to the distance relationship between the features, key features are sorted and feature subsets are selected. Finally, the feature subset is classified by using a support vector machine and adjusted parameters to achieve the best classification effect. Results By selecting key features for classification prediction, the classification accuracy of the classification model can reach 81.33%. 3-, 4-, and 5-fold cross-validation of the prediction accuracy of the support vector machine model is 77.79%~81.94%. Conclusion The pathological grading of IDC can be predicted and evaluated by texture analysis and feature extraction of breast tumors. This method can provide much valuable information for doctors' clinical diagnosis. With further development, the model demonstrates high potential for practical clinical use.

Landscape As Cultural Identity In Cau Bau Kan Movie

The aim of this research is to investigate the landscape as forming identity in the Cau Bau Kan movie scene. However, the landscape depicted is often symbolic, and often contributes to social formation, impacting on human associations and social norms. The purpose of this study is to know the meaning of landscape into a cultural identity from the Cau Bau Kan movie. This study uses texture analysis to reveal surface quality in a film. The use of textural analysis to interpret and analyze the interrelations of narrative, style, and meaning of the strategy of applying aspects of landscape as cultural identity.

Understanding and predicting sensory crispness of deep-fried battered and breaded coatings.

Crusted crispness refers to coatings with a dry and brittle surface contrasting a high‐moisture core; it is desirable for the enjoyment and quality of deep‐fried goods. This study aims to investigate instrumental measurements and sensory measurements of crispness. Deep‐fried breadcrumb coatings of eight sizes were investigated: 4.0 mm, 2.8 mm, 2.0 mm, 1.4 mm, 1.0 mm, 710 μm, 500 μm, and 355 μm. Sensory profiling was carried out to develop a tailored lexicon for deep‐fried battered and breaded shrimp. Principal component analysis highlights that large breadcrumb sizes correlate with crispness, hardness, particle size, surface color, color uniformity, surface irregularity, total porosity, maximum force, area, drop in force, number of sound peaks, and sound pressure level. Agglomerative hierarchical clustering was used to confirm clustering of samples according to breadcrumb size. Multiple factor analysis confirmed overall correlation between sensory measurements and instrumental measurements (RV = 0.810). Partial least squares regression was used to develop a predictive model for crispness from instrumental measurements (R 2 = .854). The use of texture analysis and Acoustics provide information of the structures strength and deformation behavior, while X‐ray microCT provides a high resolution and noninvasive method that acquires information on the internal morphology. These instrumental methods collectively demonstrate the relationship between microstructure to sensory. This study investigates how a change in the microstructure of deep‐fried battered and breaded coatings affect crispness perception. These changes were investigated analytically and by using a sensory panel, this is important from a manufacturing perspective in order to understand what the major contributors are to a crisp texture. The key highlights of this study include both instrumental measurements and sensory measurements can be used to measure crispness as both types of testing are correlated. Changes in the size of breadcrumbs affect both instrumental measurements and sensory measurements. A predictive model can be re‐simulated to allow prediction of crispness in deep‐fried battered and breaded coatings.

SUPPORT VECTOR MACHINE AND DECISION TREE BASED CLASSIFICATION OF SIDE-SCAN SONAR MOSAICS USING TEXTURAL FEATURES

Abstract. The diversity and heterogeneity of coastal, estuarine and stream habitats has led to them becoming a prevalent topic for study. Woody ruins are areas of potential riverbed habitat, particularly for fish. Therefore, the mapping of those areas is of interest. However, due to the limited visibility in some river systems, satellites, airborne or other camera-based systems (passive systems) cannot be used. By contrast, sidescan sonar is a popular underwater acoustic imaging system that is capable of providing high- resolution monochromatic images of the seafloor and riverbeds. Although the study of sidescan sonar imaging using supervised classification has become a prominent research subject, the use of composite texture features in machine learning classification is still limited. This study describes an investigation of the use of texture analysis and feature extraction on side-scan sonar imagery in two supervised machine learning classifications: Support Vector Machine (SVM) and Decision Tree (DT). A combination of first- order texture and second-order texture is investigated to obtain the most appropriate texture features for the image classification. SVM, using linear and Gaussian kernels along with Decision Tree classifiers, was examined using selected texture features. The results of overall accuracy and kappa coefficient revealed that SVM using a linear kernel leads to a more promising result, with 77% overall accuracy and 0.62 kappa, than SVM using either a Gaussian kernel or Decision Tree (60% and 73% overall accuracy, and 0.39 and 0.59 kappa, respectively). However, this study has demonstrated that SVM using linear and Gaussian kernels as well as a Decision Tree makes it capable of being used in side-scan sonar image classification and riverbed habitat mapping.

Using texture analysis of head CT images to differentiate osteoporosis from normal bone density

ObjectivesTo investigate the use of texture analysis for the detection of osteoporosis on noncontrast head CTs, and to explore optimal sampling regions within the craniofacial bones. MethodsIn this IRB-approved, retrospective study, the clivus, bilateral sphenoid triangles and mandibular condyles were manually segmented on each noncontrast head CT, and 41 textures features were extracted from 29 patients with normal bone density (NBD); and 29 patients with osteoporosis. Basic descriptive statistics including a false discovery rate correction were performed to evaluate for differences in texture features between the cohorts. ResulsSixteen texture features demonstrated significant differences (P < 0.01) between NBD and osteoporosis in the clivus including 4 histogram features, 2 gray-level co-occurrence matrix features, 8 gray-level run-length features and 2 Law’s features. Nineteen texture features including 9 histogram features, 1 GLCM features, 2 GLRL features, 5 Law’s features and 2 GLGM features demonstrated statistically significant differences in both sides of the sphenoid triangles. A total 24 texture features demonstrated statistically significant differences between normal BMD and osteoporosis in the left sphenoid and a total of 31 texture features in the left condyle. Furthermore, a total of 22 texture features including 6 histogram features, 3 GLCM features, 9 GLRL features, 2 Law’s features and 2 GLGM features demonstrated statistically significant differences in both sides of the mandibular condyles. ConclusionThe results of this investigation suggest that specific texture analysis features derived from regions of interest placed within multiple sites within the skull base and maxillofacial bones can distinguish between patients with normal bone mineral density compared to those with osteoporosis. This study demonstrates the potential utility of a texture analysis for identification of osteoporosis on head CT, which may help identify patients who have not undergone screening with traditional DXA.

The value of quantitative CT texture analysis in differentiation of angiomyolipoma without visible fat from clear cell renal cell carcinoma on four-phase contrast-enhanced CT images

To investigate the diagnostic performance and usefulness of texture analysis in differentiating angiomyolipoma (AML) without visible fat from clear cell renal cell carcinoma (ccRCC) on four-phase contrast-enhanced computed tomography (CECT). Seventeen patients with AML without visible fat and 50 patients with ccRCC of size ≤4.5 cm who had also undergone preoperative four-phase CECT were included in this study. The histogram, grey-level co-occurrence matrix (GLCM), and grey-level run length matrix (GLRLM) were evaluated. Sequential feature selection (SFS) and support vector machine (SVM) classifier with leave-one-out cross validation were used. Using the SFS and SVM classifiers, five texture features were selected; mean (unenhanced), standard deviation (unenhanced and excretory), cluster prominence (nephrographic), and long-run high grey-level emphasis (corticomedullary). Diagnostic performance of the five selected texture features for all CT phases was as follows: 82% sensitivity, 76% specificity, 85% accuracy, and 85 area under the receiver operating characteristic curve (AUC). In the subgroup analysis, the AUCs of each phase were significantly >0.5 (p<0.05). In the pairwise comparison of AUCs between four phases, there were no significant differences between the four phases except the unenhanced and corticomedullary phases (p=0.015), i.e., the unenhanced phase showed slightly higher AUC than the corticomedullary phase. Texture analysis of small renal masses (≤4.5 cm) on four-phase CECT can accurately differentiate AML without visible fat from ccRCC and showed good diagnostic performance for both the unenhanced and enhanced phases.

Evaluation of Early Response to Neoadjuvant Chemotherapy or Breast Cancer : Usefulness of Texture Analysis of DCE and Histogram Analysis of DW-MRI [Presidential Award Proceedings

Evaluation of Early Response to Neoadjuvant Chemotherapy or Breast Cancer : Usefulness of Texture Analysis of DCE and Histogram Analysis of DW-MRI [Presidential Award Proceedings

Tree species classification in tropical forests using visible to shortwave infrared WorldView-3 images and texture analysis

Tropical forest conservation and management can significantly benefit from information about the spatial distribution of tree species. Very-high resolution (VHR) spaceborne platforms have been hailed as a promising technology for mapping tree species over broad spatial extents. WorldView-3, the most advanced VHR sensor, provides spectral data in 16 bands covering the visible to near-infrared (VNIR, 400–1040 nm) and shortwave-infrared (SWIR, 1210–2365 nm) wavelength ranges. It also collects images at unprecedented levels of details using a panchromatic band with 0.3-m of spatial resolution. However, the potential of WorldView-3 at its full spectral and spatial resolution for tropical tree species classification remains unknown. In this study, we performed a comprehensive assessment of WorldView-3 images acquired in the dry and wet seasons for tree species discrimination in tropical semi-deciduous forests. Classification experiments were performed using VNIR individually and combined with SWIR channels. To take advantage of the sub-metric resolution of the panchromatic band for classification, we applied an individual tree crown (ITC)-based approach that employed pan-sharpened VNIR bands and gray level co-occurrence matrix texture features. We determined whether the combination of images from the two annual seasons improves the classification accuracy. Finally, we investigated which plant traits influenced species detection. The new SWIR sensing capabilities of WorldView-3 increased the average producer’s accuracy up to 7.8%, by enabling the detection of non-photosynthetic vegetation within ITCs. The combination of VNIR bands from the two annual seasons did not improve the classification results when compared to the results obtained using images from each season individually. The use of VNIR bands at their original 1.2-m spatial resolution yielded average producer’s accuracies of 43.1 ± 3.1% and 38.8 ± 3% in the wet and dry seasons, respectively. The ITC-based approach improved the accuracy to 70 ± 8% in the wet and 68.4 ± 7.4% in the dry season. Texture analysis of the panchromatic band enabled the detection of species-specific differences in crown structure, which improved species detection. The use of texture analysis, pan-sharpening, and ITC delineation is a potential approach to perform tree species classification in tropical forests with WorldView-3 satellite images.

GLRLM features in CT texture analysis are prognostic factors: Evaluation of pancreatic cancer heterogeneity, retrospective cohort study

Introduction: Pancreatic cancer is an lethal disease. Tumor heterogeneity is an important indicator of the progression of chemoresistance. Although texture analyses can allow quantification of tumor heterogeneity, only a few studies reported the use of texture analysis in cases of pancreatic cancer. Therefore, we analyzed tumor heterogeneity in preoperative computed tomography (CT) scans by performing texture analysis using the gray-level run-length matrix (GLRLM). Method: We analyzed 116 consecutive patients who underwent surgical resection during 2001-2017 and had preoperative CT data available for analysis. An ROI was drawn on slices with a visible tumor and normal pancreas on the arterial phase CT scans and the correlation of pathological characteristics and GLRLM features was analyzed. We then performed Kaplan—Meier survival curve analysis among pancreatic cancer patients. Result: The gray-level non-uniformity (GLN) values in GLRLM features for tumors were higher than those for normal pancreas. High GLN values represent a non-uniform texture, i.e., heterogeneity. Recurrence-free survival was shorter in the group with high GLN135 values (p = 0.025). Conclusion: Tumor heterogeneity is a significant prognostic factor. Our analyses of the correlation between pathological outcomes and GLRLM features in pancreatic cancer patients showed that GLN values were a powerful indicator of prognosis.