Haralick Texture Features Research Articles

Paddy seed variety identification using T20-HOG and Haralick textural features

The seed is an inevitable element for agricultural and industrial production. The non-destructive paddy seed variety identification is essential to assure paddy purity and quality. This research is aimed at developing a computer vision-based system to identify paddy varieties using multiple heterogeneous features, exploiting textural, external, and physical properties. We captured the paddy seed images without any fixed setup to make the system user friendly at both industry and farmer levels, which can lead to illumination problems in the images. To overcome this problem, we introduced a modified histogram oriented gradient (T20-HOG) feature that can describe the illumination, scale, and rotational variations of a paddy image. We also utilized the existing Haralick and traditional features and the dimensionality of the features is reduced by the Lasso feature selection technique. The selected features are used to train the feed-forward neural network (FNN) to predict the paddy variety. The experiments conducted on two different datasets: BDRICE, and VNRICE. Results of our method are shown in terms of four standard evaluation metrics, namely, accuracy, precision, recall, and F_1 score, and achieved 99.28%, 98.64%, 98.48%, and 98.56% score, respectively. We also compared our system efficiency with existing studies. The experimental results demonstrate that our proposed features are effective to identify paddy variety and achieved a new state-of-the-art performance. And we also observed that our newly proposed T20-HOG features have a major impact on overall system performance.

Web-based remote sensing image retrieval using multiscale and multidirectional analysis based on Contourlet and Haralick texture features

PurposeExtracting suitable features to represent an image based on its content is a very tedious task. Especially in remote sensing we have high-resolution images with a variety of objects on the Earth's surface. Mahalanobis distance metric is used to measure the similarity between query and database images. The low distance obtained image is indexed at the top as high relevant information to the query.Design/methodology/approachThis paper aims to develop an automatic feature extraction system for remote sensing image data. Haralick texture features based on Contourlet transform are fused with statistical features extracted from the QuadTree (QT) decomposition are developed as feature set to represent the input data. The extracted features will retrieve similar images from the large image datasets using an image-based query through the web-based user interface.FindingsThe developed retrieval system performance has been analyzed using precision and recall and F1 score. The proposed feature vector gives better performance with 0.69 precision for the top 50 relevant retrieved results over other existing multiscale-based feature extraction methods.Originality/valueThe main contribution of this paper is developing a texture feature vector in a multiscale domain by combining the Haralick texture properties in the Contourlet domain and Statistical features using QT decomposition. The features required to represent the image is 207 which is very less dimension compare to other texture methods. The performance shows superior than the other state of art methods.

Haralick texture features extracted from Ga-68 PSMA PET/CT to differentiate normal prostate from prostate cancer: a feasibility study.

Role of texture parameters on the basis of Ga-68 PSMA PET/CT in prostate cancer (Pca) is largely unexplored. Present work done is a preliminary study that aims to evaluate the role of Haralick texture features on the basis of Ga-68 PSMA PET/CT in Pca in which texture features were used to differentiate between normal prostate and Pca tissue. The study retrospectively enrolled patients in two groups: group 1 included 30 patients with biopsy-proven adenocarcinoma prostate and median age 64 years (range: 50-82 years) who underwent baseline Ga-68 PSMA PET/CT prior to therapy; group 2 included 24 patients with pathologies other than Pca and median age 53.5 years (range: 18-80 years) who underwent Ga-68 PSMA PET/CT as part of another study in our department. Patients in group 2 did not have any prostate pathology and served as controls for the study. The segmented images of prostate (3-D image) were used to calculate 11 Haralick texture features in MATLAB. SUVmax was also evaluated. All parameters were compared among the two groups using appropriate statistical analysis and P value <0.05 was considered significant. All 11 Haralick texture features, as well as SUVmax, were significantly different among Pca and controls (P < 0.05). Among the texture features, contrast was most significant (P value of Mann-Whitney U <0.001) in differentiating Pca from normal prostate with AUROC curve of 82.9% with sensitivity and specificity 83.30% and 73.30%, respectively at cut-off 0.640. SUVmax was also significant with AUROC curve 94.0% and sensitivity and specificity 62.5% and 90%, respectively at cut-off 5.7. A significant negative correlation of SUVmax was observed with contrast. Haralick texture features have a significant role in differentiating Pca and normal prostate.

Recognition of crevasses with high‐resolution digital elevation models: Application of geomorphometric modeling and texture analysis

AbstractCrevasses—cracks in glaciers and ice sheets—pose a danger to polar researchers and glaciologists. We compare the capabilities of two techniques—geomorphometric modeling and texture analysis—to recognize open and hidden crevasses using high‐resolution digital elevation models (DEMs) generated from images collected by an unmanned aerial system (UAS). The first technique includes derivation of local morphometric variables; the second includes calculation of the Haralick texture features. The study area is represented by the first 30 km of a sledge route between the Progress and Vostok polar stations, East Antarctica. The UAS survey was performed by a Geoscan 201 Geodesy UAS. For the sledge route area, DEMs with resolutions of 0.25, 0.5, and 1 m were generated. Models of 12 morphometric variables and 11 texture features were derived from the DEMs. In terms of crevasse recognition, the most informative morphometric variable and texture feature was horizontal curvature and inverse difference moment, respectively. In most cases, derivation and mapping of these variables allow one to recognize crevasses wider than 3 m; narrower crevasses can be recognized for lengths from 500 m. For crevasse recognition, the geomorphometric modeling and the Haralick texture analysis can complement each other.

Deep Learning in Classification of Covid-19 Coronavirus, Pneumonia and Healthy Lungs on CXR and CT Images.

PurposeIn this paper, the transfer learning method has been implemented to chest X-ray (CXR) and computed tomography (CT) bio-images of diverse kinds of lungs maladies, including CORONAVIRUS 2019 (COVID-19). COVID-19 identification is a difficult assignment that constantly demands a careful analysis of a patient’s clinical images, as COVID-19 is found to be very alike to pneumonic viral lung infection. In this paper, a transfer learning model to accelerate prediction processes and to assist medical professionals is proposed. Finally, the main purpose is to do an accurate classification between Covid-19, pneumonia and, healthy lungs using CXR and CT images.MethodsLearning transfer gives the possibility to find out about this new illness COVID-19, using the knowledge we have about the pneumonia virus. This demonstrates the apprehensiveness achieved from a new architecture trained to detect virus-related pneumonia that must be transferred for COVID-19 detection. Transfer learning presents a considerable dissimilarity in results when compared to the result of traditional groupings. It is not necessary to create a separate model for the classification of COVID-19. This simplifies complicated issues by adopting the available model for COVID-19 determination. Automated diagnosis of COVID-19 using Haralick texture features is focused on segmented lung images and problematic lung patches. Lung patches are necessary for the augmentation of COVID-19 image data.ResultsThe obtained outcomes are quite reliable for all distinctive processes as the proposed architecture can distinguish healthy lungs, pneumonia, COVID-19.ConclusionsThe results suggest that the implemented model is improved considering other existing models because the obtained classification accuracy is over the recently obtained results. It is a belief that the new architecture that is implemented in this study, delivers a petite step in building refined Coronavirus 2019 diagnosis architecture using CXR and CT bio-images.

Computer Aided Breast Cancer Detection Using Ensembling of Texture and Statistical Image Features.

Breast cancer, like most forms of cancer, is a fatal disease that claims more than half a million lives every year. In 2020, breast cancer overtook lung cancer as the most commonly diagnosed form of cancer. Though extremely deadly, the survival rate and longevity increase substantially with early detection and diagnosis. The treatment protocol also varies with the stage of breast cancer. Diagnosis is typically done using histopathological slides from which it is possible to determine whether the tissue is in the Ductal Carcinoma In Situ (DCIS) stage, in which the cancerous cells have not spread into the encompassing breast tissue, or in the Invasive Ductal Carcinoma (IDC) stage, wherein the cells have penetrated into the neighboring tissues. IDC detection is extremely time-consuming and challenging for physicians. Hence, this can be modeled as an image classification task where pattern recognition and machine learning can be used to aid doctors and medical practitioners in making such crucial decisions. In the present paper, we use an IDC Breast Cancer dataset that contains 277,524 images (with 78,786 IDC positive images and 198,738 IDC negative images) to classify the images into IDC(+) and IDC(-). To that end, we use feature extractors, including textural features, such as SIFT, SURF and ORB, and statistical features, such as Haralick texture features. These features are then combined to yield a dataset of 782 features. These features are ensembled by stacking using various Machine Learning classifiers, such as Random Forest, Extra Trees, XGBoost, AdaBoost, CatBoost and Multi Layer Perceptron followed by feature selection using Pearson Correlation Coefficient to yield a dataset with four features that are then used for classification. From our experimental results, we found that CatBoost yielded the highest accuracy (92.55%), which is at par with other state-of-the-art results—most of which employ Deep Learning architectures. The source code is available in the GitHub repository.

Evaluation of statistical and Haralick texture features for lymphoma histological images classification

ABSTRACT The investigation of different types of cancer can be performed by images classification with features extracted from specific regions identified by a segmentation step. Therefore, this study presents the evaluation of texture features extracted from neoplastic nuclei for the classification of lymphomas images. The neoplastic nuclei were segmented by steps of pre and post-processing and a thresholding. Statistical and Haralick’s features extracted from wavelet and ranklet transforms were evaluated with different classifiers. The use of the statistical metrics from the wavelet transform in association with the K-nearest neighbour classifier provided the best results in most of the two-class classifications.

Remote sensing image retrieval by integrating automated deep feature extraction and handcrafted features using curvelet transform

Deep learning techniques have become increasingly popular for classifying large-scale image and video data. Remote sensing applications require robust search engines to retrieve similar information dependent on an example-based query instead of a tag-based query. Deep features can be extracted automatically by training raw data without having any domain-specific knowledge. However, the training time for a massive amount of multimedia datasets is high. Training complexity is reduced using pre-trained GoogleNet weights for initial feature extraction. To fine-tune the feature vector and reduce the dimensionality, a one dimension convolutional neural network (1D-CNN) is applied. There is a loss of information while resizing the input image to a pre-trained network with an acceptable input size. We proposed a new feature set by integrating handcrafted features at detailed scales and deep features to improve the system’s efficiency. The curvelet transform was used to decompose the image into coarse and detailed scales. Haralick texture features were extracted from the detail coefficients in four directions and fused with fine-tuned deep features. The proposed feature set was assessed using standard performance metrics from the literature. The proposed technique achieved improved performance with 89% accuracy for retrieval of the top 50 relevant results.

Detection of Breast Cancer on Magnetic Resonance Imaging Using Hybrid Feature Extraction and Deep Neural Network Techniques

Breast cancer is one of the most occurring cancers in women due to the uncontrolled growth of abnormal cells in the lobules or milk ducts. The treatment for the breast cancer at an early stage is important using Magnetic Resonance Imaging (MRI) which effectively measures the size of the cancer and also checks tumors in the opposite breast. The deposition of calcium components on the breast tissue is known as micro-calcifications. The calcium salts deposited in the breast are involved with the cancer and were not diagnosed accurately due to the low effectiveness of existing imaging technique namely Haralick feature extraction technique. The MRI breast cancer diagnosis creates problems during classification of breast image and leads to misclassifications, such as unidentified calcium deposits in the existing K-Nearest Neighbour (KNN) classifier. The misclassification issues are overcome by an accurate classification and identification of calcium salts and checks whether deposited salt on breast tissue is involved with cancer or not. Initially, Contrast-Limited Adaptive Histogram Equalization (CLAHE) is used to remove the unwanted noise in the MRI and Morphological, Multilevel Otsu’s Thresholding and region growing techniques perform segmentation to mask unwanted breast tissues. The proposed Hybrid LOOP Haralick feature extraction technique is developed by combining the both Local Optimal Oriented Pattern (LOOP) and Haralick texture feature and the hybrid parameters are applied to the Stacked Auto Encoder based (SAE) to classify the breast MRI image as a Malignant or Benign. The performance of the proposed hybrid LOOP Haralick feature extraction shows significant accuracy improvement of 3.83% when compared to the Haralick feature extraction technique.

An Efficient Gait Recognition Method for Known and Unknown Covariate Conditions

Gait is a unique non-invasive biometric form that can be utilized to effectively recognize persons, even when they prove to be uncooperative. Computer-aided gait recognition systems usually use image sequences without considering covariates like clothing and possessions of carrier bags whilst on the move. Similarly, in gait recognition, there may exist unknown covariate conditions that may affect the training and testing conditions for a given individual. Consequently, common techniques for gait recognition and measurement require a degree of intervention leading to the introduction of unknown covariate conditions, and hence this significantly limits the practical use of the present gait recognition and analysis systems. To overcome these key issues, we propose a method of gait analysis accounting for both known and unknown covariate conditions. For this purpose, we propose two methods, i.e., a Convolutional Neural Network (CNN) based gait recognition and a discriminative features-based classification method for unknown covariate conditions. The first method can handle known covariate conditions efficiently while the second method focuses on identifying and selecting unique covariate invariant features from the gallery and probe sequences. The feature set utilized here includes Local Binary Patterns (LBP), Histogram of Oriented Gradients (HOG), and Haralick texture features. Furthermore, we utilize the Fisher Linear Discriminant Analysis for dimensionality reduction and selecting the most discriminant features. Three classifiers, namely Random Forest, Support Vector Machine (SVM), and Multilayer Perceptron are used for gait recognition under strict unknown covariate conditions. We evaluated our results using CASIA and OUR-ISIR datasets for both clothing and speed variations. As a result, we report that on average we obtain an accuracy of 90.32% for the CASIA dataset with unknown covariates and similarly performed excellently on the ISIR dataset. Therefore, our proposed method outperforms existing methods for gait recognition under known and unknown covariate conditions.

CLASSIFICATION OF BRAIN MRI IMAGES BY USING THE AUTOMATIC SEGMENTATION AND TEXTURE ANALYSIS

Brain tumor is a relatively severe human disease type. Its timely diagnosis and tumor type definition are an actual task in modern medicine. Lately, the segmentation methods on 3D brain images (like computer and magnetic resonance tomography) are used for definition of a certain tumor type. Nevertheless, the segmentation is usually conducted manually, which requires a lot of time and depends on the experience of a doctor. This paper looks at the possibility of creating a method for the automatic segmentation of images. As a training sample, the medical database of MRI brain tomography with three tumor types (meningioma, glioma, and pituitary tumor) was taken. Taking into account the different slices, the base had: 708 examples of meningioma, 1426 examples of glioma, and 930 examples of pituitary tumor. The database authors marked the regions of interest on each image, which were used as a tutor (supervised learning) for automatic segmentation model. Before model creation, currently existing popular automatic segmentation models were analyzed. U-Net deep convolution neural network architecture was used as the most suitable one. As the result of its use, the model was obtained, which can segment the image correctly in seventy four percent of six hundred images (testing sample). After obtaining the automatic segmentation model, the Random Forest models for three “One versus All”tasks and one multiclass task were created for brain tumor classification. The total sample was divided into training (70 %), testing (20 %), and examining (10 %) ones before creating the models. The accuracy of the models in the examining sample varies from 84to 94 percent. For model classification creation, the texture features were used, obtained by texture analysis method, and created by the co-authors of the Department of Biomedical Cybernetics in the task of liver ultrasound image classification. They were compared with well-known Haralick texture features. The comparison showed that the best way to achieve an accurate classification model is to combine all the features into one stack.

Land-Use and Land-Cover Classification Using a Human Group-Based Particle Swarm Optimization Algorithm with an LSTM Classifier on Hybrid Pre-Processing Remote-Sensing Images

Land-use and land-cover (LULC) classification using remote sensing imagery plays a vital role in many environment modeling and land-use inventories. In this study, a hybrid feature optimization algorithm along with a deep learning classifier is proposed to improve the performance of LULC classification, helping to predict wildlife habitat, deteriorating environmental quality, haphazard elements, etc. LULC classification is assessed using Sat 4, Sat 6 and Eurosat datasets. After the selection of remote-sensing images, normalization and histogram equalization methods are used to improve the quality of the images. Then, a hybrid optimization is accomplished by using the local Gabor binary pattern histogram sequence (LGBPHS), the histogram of oriented gradient (HOG) and Haralick texture features, for the feature extraction from the selected images. The benefits of this hybrid optimization are a high discriminative power and invariance to color and grayscale images. Next, a human group-based particle swarm optimization (PSO) algorithm is applied to select the optimal features, whose benefits are a fast convergence rate and ease of implementation. After selecting the optimal feature values, a long short-term memory (LSTM) network is utilized to classify the LULC classes. Experimental results showed that the human group-based PSO algorithm with a LSTM classifier effectively well differentiates the LULC classes in terms of classification accuracy, recall and precision. A maximum improvement of 6.03% on Sat 4 and 7.17% on Sat 6 in LULC classification is reached when the proposed human group-based PSO with LSTM is compared to individual LSTM, PSO with LSTM, and Human Group Optimization (HGO) with LSTM. Moreover, an improvement of 2.56% in accuracy is achieved, compared to the existing models, GoogleNet, Visual Geometric Group (VGG), AlexNet, ConvNet, when the proposed method is applied.

Tumor Habitat-derived Radiomic Features at Pretreatment MRI That Are Prognostic for Progression-free Survival in Glioblastoma Are Associated with Key Morphologic Attributes at Histopathologic Examination: A Feasibility Study.

To identify radiomic features extracted from the tumor habitat on routine MR images that are prognostic for progression-free survival (PFS) and to assess their morphologic basis with corresponding histopathologic attributes in glioblastoma (GBM). In this retrospective study, 156 pretreatment GBM MR images (gadolinium-enhanced T1-weighted, T2-weighted, and fluid-attenuated inversion recovery [FLAIR] images) were curated. Of these 156 images, 122 were used for training (90 from The Cancer Imaging Archive and 32 from the Cleveland Clinic, acquired between December 1, 2011, and May 1, 2018) and 34 were used for validation. The validation set was obtained from the Ivy Glioblastoma Atlas Project database, for which the percentage extent of 11 histologic attributes was available on corresponding histopathologic specimens of the resected tumor. Following expert annotations of the tumor habitat (necrotic core, enhancing tumor, and FLAIR-hyperintense subcompartments), 1008 radiomic descriptors (eg, Haralick texture features, Laws energy features, co-occurrence of local anisotropic gradient orientations [CoLIAGe]) were extracted from the three MRI sequences. The top radiomic features were obtained from each subcompartment in the training set on the basis of their ability to risk-stratify patients according to PFS. These features were then concatenated to create a radiomics risk score (RRS). The RRS was independently validated on a holdout set. In addition, correlations (P < .05) of RRS features were computed, with the percentage extent of the 11 histopathologic attributes, using Spearman correlation analysis. RRS yielded a concordance index of 0.80 on the validation set and constituted radiomic features, including Laws (capture edges, waves, ripple patterns) and CoLIAGe (capture disease heterogeneity) from enhancing tumor and FLAIR hyperintensity. These radiomic features were correlated with histopathologic attributes associated with disease aggressiveness in GBM, particularly tumor infiltration (P = .0044) and hyperplastic blood vessels (P = .0005). Preliminary findings demonstrated significant associations of prognostic radiomic features with disease-specific histologic attributes, with implications for risk-stratifying patients with GBM for personalized treatment decisions. Supplemental material is available for this article. © RSNA, 2020.

A Computer Vision Based System for Classification of Chemically and Naturally Ripened Mangoes

Recently there was news indicating that mangoes might cause cancer. The news was based on the fact that mangoes were being artificially ripened using a chemical- calcium carbide and Ethrel, a well- known carcinogenic. The consumers hence have to be careful in buying the mangoes. In this study, we have proposed a model for classification of artificially and naturally ripened mangoes using k-NN and SVM classifiers. In order to improve the efficacy of the model, color space features such like RGB, HSV, L*a*b are extracted. Along with the color space features, 14 Haralick texture features are also extracted here. An mango is automatically segmented in an image using modified K-means clustering segmentation method. For the experimental study, mangoes of 2 varieties such as Badami and Raspuri have been taken. In each variety, three different classes of ripened mangoes are taken such as naturally and in chemical, two artificial ripening treatments were applied like calcium carbide and Ethrel solution. The obtained experimental result in terms of F-measure is ranging from 64% to 84% for two different varieties of mangoes using two different chemicals. Further this proposed model can be implemented for different variety of mangoes.

Texture Analysis of F-18 Fluciclovine PET/CT to Predict Biochemically Recurrent Prostate Cancer: Initial Results.

Predicting biochemical recurrence of prostate cancer is imperative for initiating early treatment, which can improve the outcome of cancer treatment. However, because of inter- and intrareader variability in interpretation of F-18 fluciclovine positron emission tomography/computed tomography (PET/CT), it is difficult to reliably discern between necrotic tissue owing to radiation therapy and tumor tissue. Our goal is to develop a computational methodology using Haralick texture analysis that can be used as an adjunct tool to improve and standardize the interpretation of F-18 fluciclovine PET/CT to identify biochemical recurrence of prostate cancer. Four main textural features were chosen by variable selection procedure using least absolute shrinkage and selection operator logistic regression and bootstrapping, and then included as predictors in subsequent logistic ridge regression model for prediction (n = 28). Age at prostatectomy, prostate-specific antigen (PSA) level before the PET/CT imaging, and number of days between the prostate-specific antigen measurement and PET/CT imaging were also included in the prediction model. The overfitting-corrected area under the curve and Brier score of the proposed model were 0.94 (95% CI: 0.81, 1.00) and 0.12 (95% CI: 0.03, 0.23), respectively. Compared with a model with textural features (TI model) and that with only clinical information (CI model), the proposed model achieved 2% and 32% increase in AUC and 8% and 48% reduction in Brier score, respectively. Combining Haralick textural features based on the PET/CT imaging data with clinical information shows a high potential of enhanced prediction of the biochemical recurrence of prostate cancer.

Classification of sea ice types in Sentinel-1 synthetic aperture radar images

Abstract. A new Sentinel-1 image-based sea ice classification algorithm using a machine-learning-based model trained in a semi-automated manner is proposed to support daily ice charting. Previous studies mostly rely on manual work in selecting training and validation data. We show that the readily available ice charts from the operational ice services can reduce the amount of manual work in preparation of large amounts of training/testing data. Furthermore, they can feed highly reliable data to the trainer by indirectly exploiting the best ability of the sea ice experts working at the operational ice services. The proposed scheme has two phases: training and operational. Both phases start from the removal of thermal, scalloping, and textural noise from Sentinel-1 data and calculation of grey level co-occurrence matrix and Haralick texture features in a sliding window. In the training phase, the weekly ice charts are reprojected into the SAR image geometry. A random forest classifier is trained with the texture features on input and labels from the rasterized ice charts on output. Then, the trained classifier is directly applied to the texture features from Sentinel-1 images operationally. Test results from the two datasets spanning winter (January–March) and summer (June–August) seasons acquired over the Fram Strait and the Barents Sea showed that the classifier is capable of retrieving three generalized cover types (open water, mixed first-year ice, old ice) with overall accuracies of 87 % and 67 % in winter and summer seasons, respectively. For the summer season, the classifier failed in distinguishing mixed first-year ice from old ice with accuracy of only 12 %; however, it performed rather like an ice–water discriminator with high accuracy of 98 % as the misclassification between the mixed first-year ice and old ice was between them. The accuracy for five cover types (open water, new ice, young ice, first-year ice, old ice) in the winter season was 60 %. The errors are attributed both to incorrect manual classification on the ice charts and to the semi-automated algorithm. Finally, we demonstrate the potential for near-real-time service of the ice map using daily mosaicked Sentinel-1 images.

Comparison of 1.5 T and 3 T MRI hippocampus texture features in the assessment of Alzheimer's disease

ObjectiveMany studies evaluated how the Magnetic Resonance Imaging (MRI) field strength affects the effectiveness to detect neurodegenerative changes of Alzheimer's disease (AD), derived from atrophy or thickness. To the best of our knowledge, no study evaluated before how tissue texture changes are affected. In this research, hippocampus texture features extracted from 1.5 T and 3 T MRI are evaluated how are affected by the magnetic field strength. MethodsMR imaging data from 14 Normal Controls (NC), 14 with Mild Cognitive Impairment (MCI), 11 MCI converters (MCIc) and 10 AD subjects scanned at 1.5 T and 3 T were included. Haralick's texture features were extracted from the hippocampus, along with hippocampal and amygdala volumes and cortical thickness. One-way ANOVA, paired-samples and Wilcoxon signed t-tests were used to evaluate if there were significant differences between the features. Results3 T texture features were significantly different for NC vs AD, NC vs MCI and MCI vs AD, whereas, 1.5 T for MCI vs AD only. Amygdala and hippocampal volumes, showed significant differences for NC vs AD for both MRI strengths, whereas cortical thickness for MCI vs MCIc for the 3 T. Paired sample t-test and Wilcoxon signed-rank test revealed significant differences for Angular Second Moment (ASM), contrast, correlation, variance, sum variance and entropy, the amygdala volume and cortical thickness. Between NC vs MCI, 3 T texture revealed higher Area Under Curve (AUC). Conclusion3 T texture revealed significant differences for more features compared to 1.5 T, whereas, atrophy and thickness had similar results. Significance3 T texture changes provide earlier diagnosis compared to 1.5 T volume or texture changes.

Machine learning assisted DSC-MRI radiomics as a tool for glioma classification by grade and mutation status

BackgroundCombining MRI techniques with machine learning methodology is rapidly gaining attention as a promising method for staging of brain gliomas. This study assesses the diagnostic value of such a framework applied to dynamic susceptibility contrast (DSC)-MRI in classifying treatment-naïve gliomas from a multi-center patients into WHO grades II-IV and across their isocitrate dehydrogenase (IDH) mutation status.MethodsThree hundred thirty-three patients from 6 tertiary centres, diagnosed histologically and molecularly with primary gliomas (IDH-mutant = 151 or IDH-wildtype = 182) were retrospectively identified. Raw DSC-MRI data was post-processed for normalised leakage-corrected relative cerebral blood volume (rCBV) maps. Shape, intensity distribution (histogram) and rotational invariant Haralick texture features over the tumour mask were extracted. Differences in extracted features across glioma grades and mutation status were tested using the Wilcoxon two-sample test. A random-forest algorithm was employed (2-fold cross-validation, 250 repeats) to predict grades or mutation status using the extracted features.ResultsShape, distribution and texture features showed significant differences across mutation status. WHO grade II-III differentiation was mostly driven by shape features while texture and intensity feature were more relevant for the III-IV separation. Increased number of features became significant when differentiating grades further apart from one another. Gliomas were correctly stratified by mutation status in 71% and by grade in 53% of the cases (87% of the gliomas grades predicted with distance less than 1).ConclusionsDespite large heterogeneity in the multi-center dataset, machine learning assisted DSC-MRI radiomics hold potential to address the inherent variability and presents a promising approach for non-invasive glioma molecular subtyping and grading.

Outcomes assessment in intrahepatic cholangiocarcinoma using qualitative and quantitative imaging features

BackgroundTo assess the performance of imaging features, including radiomics texture features, in predicting histopathologic tumor grade, AJCC stage, and outcomes [time to recurrence (TTR) and overall survival (OS)] in patients with intrahepatic cholangiocarcinoma (ICC).MethodsSeventy-three patients (26 M/47F, mean age 63y) with pre-operative imaging (CT, n = 37; MRI, n = 21; CT and MRI, n = 15] within 6 months of resection were included in this retrospective study. Qualitative imaging traits were assessed by 2 observers. A 3rd observer measured tumor apparent diffusion coefficient (ADC), enhancement ratios (ERs), and Haralick texture features. Blood biomarkers and imaging features were compared with histopathology (tumor grade and AJCC stage) and outcomes (TTR and OS) using log-rank, generalized Wilcoxon, Cox proportional hazards regression, and Fisher exact tests.ResultsMedian TTR and OS were 53.9 and 79.7 months. ICC recurred in 64.4% (47/73) of patients and 46.6% (34/73) of patients died. There was fair accuracy for some qualitative imaging features in the prediction of worse tumor grade (maximal AUC of 0.68 for biliary obstruction on MRI, p = 0.032, observer 1) and higher AJCC stage (maximal AUC of 0.73 for biliary obstruction on CT, p = 0.002, observer 2; and AUC of 0.73 for vascular involvement on MRI, p = 0.01, observer 2). Cox proportional hazards regression analysis showed that CA 19–9 [hazard ratio (HR) 2.44/95% confidence interval (CI) 1.31–4.57/p = 0.005)] and tumor size on imaging (HR 1.13/95% CI 1.04–1.22/p = 0.003) were significant predictors of TTR, while CA 19–9 (HR 4.08/95% CI 1.75–9.56, p = 0.001) and presence of metastatic lymph nodes at histopathology (HR 2.86/95% CI 1.35–6.07/p = 0.006) were significant predictors of OS. On multivariable analysis, satellite lesions on CT (HR 2.79/95%CI 1.01–7.15/p = 0.032, observer 2), vascular involvement on MRI (HR 0.10/95% CI 0.01–0.85/p = 0.032, observer 1), and texture feature MRI variance (HR 0.55/95% CI 0.31–0.97, p = 0.040) predicted TTR once adjusted for the independent predictors CA 19–9 and tumor size on imaging. Several qualitative and quantitative features demonstrated associations with TTR, OS, and AJCC stage at univariable analysis (range: HR 0.35–19; p < 0.001–0.045), however none were predictive of OS at multivariable analysis when adjusted for CA 19–9 and metastatic lymph nodes (p > 0.088).ConclusionsThere was reasonable accuracy in predicting tumor grade and higher AJCC stage in ICC utilizing certain qualitative and quantitative imaging traits. Serum CA 19–9, tumor size, presence of metastatic lymph nodes, and qualitative imaging traits of satellite lesions and vascular involvement are predictors of patient outcomes, along with a promising predictive ability of certain quantitative texture features.

A computer aided diagnostic method for the evaluation of type II diabetes mellitus in facial thermograms.

Almost 50% of individuals around the globe are unaware of diabetes and its complications. So, an early screening of diabetes is very important at this current situation. To overcome the difficulties such as pain and discomfort to the subjects obtained from the biochemical diagnostic procedures; an infrared thermography is the diagnostic technique which measures the skin surface temperature noninvasively. Thus, the aim of our proposed study was to evaluate the type II diabetes in facial thermograms and to develop a computer aided diagnosis (CAD) system to classify the normal and diabetes. The facial thermograms (n = 160) including male (n = 79) and female (n = 81) were captured using FLIR A 305sc infrared thermal camera. The Haralick textural features were extracted from the facial thermograms based on gray level co-occurrence matrix algorithm. The TROI, TMAX, and TTOT are the statistical temperature parameters exhibited a significant negative correlation with HbA1c (r = - 0.421, - 0.411, - 0.242, p < 0.01 (TROI); r = - 0.259, p < 0.01(TMAX) and - 0.173, p < 0.05 (TTOT)). An optimal regression equation has been constructed by using the significant facial variables and standard HbA1c values. The model has achieved sensitivity, specificity, and accuracy rate as 91.42%, 88.57%, and 90% respectively. The anthropometrical variables, extracted textural features and temperature parameters were fed into the classifiers and their performances were compared. The Support Vector Machine outperformed the Linear Discriminant Analysis (84.37%) and k-Nearest Neighbor (81.25%) classifiers with the maximum accuracy rate of 89.37%. The developed CAD system has achieved 89.37% of accuracy rate for the classification of diabetes. Thus, the facial thermography could be used as the basic non-invasive prognostic tool for the evaluation of type II diabetes mellitus.