Search Results (412)

High uniformity of positron emission tomography (PET) images in the field of nuclear medicine is necessary to obtain excellent and stable data from the system. In this study, we aimed to apply and optimize a PET/magnetic resonance (MR) imaging system by approaching the gray-level co-occurrence matrix (GLCM), which is known to be efficient in the uniformity correction of images. CAIPIRINHA Dixon-VIBE was used as an MR image acquisition pulse sequence for the fast and accurate attenuation correction of PET images, and the phantom was constructed by injecting NaCl and NaCl + NiSO₄ solutions. The lambda value of the GLCM algorithm for uniformity correction of the acquired PET images was optimized in terms of energy and contrast. By applying the GLCM algorithm optimized in terms of energy and contrast to the PET images of phantoms using NaCl and NaCl + NiSO₄ solutions, average percent image uniformity (PIU) values of 26.01 and 83.76 were derived, respectively. Compared to the original PET image, an improved PIU value of more than 30% was derived from the PET image to which the proposed optimized GLCM algorithm was applied. In conclusion, we demonstrated that an algorithm optimized in terms of the GLCM energy and contrast can improve the uniformity of PET images. Full article

The authentication of Ziziphi Spinosae Semen (ZSS), Ziziphi Mauritianae Semen (ZMS), and Hovenia Acerba Semen (HAS) has become challenging. The chromatic and textural properties of ZSS, ZMS, and HAS are analyzed in this study. Color features were extracted via RGB, CIELAB, and HSI spaces, whereas texture information was analyzed via the gray-level co-occurrence matrix (GLCM) and Law’s texture feature analysis. The results revealed significant differences in color and texture among the samples. The fire–ice ion dimensionality reduction algorithm effectively fuses these features, enhancing their differentiation ability. Principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA) confirmed the algorithm’s effectiveness, with variable importance in projection analysis (VIP analysis) (VIP > 1, p < 0.05) highlighting significant differences, particularly for the fire value, which is a key factor. To further validate the reliability of the algorithm, Back Propagation Neural Network (BP), Support Vector Machine (SVM), Deep Belief Network (DBN), and Random Forest (RF) were used for reverse validation, and the accuracy of the training set and test set reached 98.83–100% and 95.89–99.32%, respectively. The method provides a simple, low-cost, and high-precision tool for the fast and nondestructive detection of food authenticity. Full article

Background: This study aimed to explore the differences in quantitative diffusion-weighted (DW) MRI parameters in oropharyngeal squamous cell carcinoma (OPC) based on Human Papillomavirus (HPV) status before and during radiotherapy (RT). Methods: Echo planar DW sequences acquired before and during (chemo)radiotherapy (CRT) of 178 patients with histologically proven OPC were prospectively analyzed. The volumetric region of interest (ROI) was manually drawn on the apparent diffusion coefficient (ADC) map, and 105 DW-MRI radiomic parameters were extracted. Change in ADC values (Δ ADC) was calculated as the difference between baseline and during RT at week 4, normalized by the baseline values. Results: Pre-treatment first-order 10th percentile ADC and Gray Level co-occurrence matrix (GLCM)-correlation were significantly lower in HPV-positive compared with HPV-negative tumors (82.4 × 10⁻⁵ mm²/s vs. 90.3 × 10⁻⁵ mm²/s, p = 0.03 and 0.18 vs. 0.30, p < 0.01). In the fourth week of RT, all first-order ADC values were significantly higher in HPV-positive tumors (p < 0.01). Δ ADC mean was significantly higher for the HPV-positive compared with the HPV-negative OPC group (95% vs. 55%, p < 0.01). A predictive model for HPV status based on smoking status, alcohol consumption, GLCM correlation, and mean ADC and 10th percentile ADC values yielded an area under the curve of 0.77 (95% CI 0.70–0.84). Conclusions: Our results highlight the potential of DW-MR imaging as a non-invasive biomarker for the prediction of HPV status, although its current role remains supplementary to pathological confirmation. Full article

Determining the maturity of cocoa pods early is not just about guaranteeing harvest quality and optimizing yield. It is also about efficient resource management. Rapid identification of the stage of maturity helps avoid losses linked to a premature or late harvest, improving productivity. Early determination of cocoa pod maturity ensures both the quality and quantity of the harvest, as immature or overripe pods cannot produce premium cocoa beans. Our innovative research harnesses artificial intelligence and computer vision technologies to revolutionize the cocoa industry, offering precise and advanced tools for accurately assessing cocoa pod maturity. Providing an objective and rapid assessment enables farmers to make informed decisions about the optimal time to harvest, helping to maximize the yield of their plantations. Furthermore, by automating this process, these technologies reduce the margins for human error and improve the management of agricultural resources. With this in mind, our study proposes to exploit a computer vision method based on the GLCM (gray level co-occurrence matrix) algorithm to extract the characteristics of images in the RGB (red, green, blue) and LAB (luminance, axis between red and green, axis between yellow and blue) color spaces. This approach allows for in-depth image analysis, which is essential for capturing the nuances of cocoa pod maturity. Next, we apply classification algorithms to identify the best performers. These algorithms are then combined via stacking and voting techniques, allowing our model to be optimized by taking advantage of the strengths of each method, thus guaranteeing more robust and precise results. The results demonstrated that the combination of algorithms produced superior performance, especially in the LAB color space, where voting scored 98.49% and stacking 98.71%. In comparison, in the RGB color space, voting scored 96.59% and stacking 97.06%. These results surpass those generally reported in the literature, showing the increased effectiveness of combined approaches in improving the accuracy of classification models. This highlights the importance of exploring ensemble techniques to maximize performance in complex contexts such as cocoa pod maturity classification. Full article

Background: Lung cancer still maintains the leading position among causes of death in the world; the process of early detection surely contributes to changes in the survival of patients. Standard diagnostic methods are grossly insensitive, especially in the early stages. In this paper, radiomic features are discussed that can assure improved diagnostic accuracy through automated lung cancer detection by considering the important feature categories, such as texture, shape, and intensity, originating from the CT DICOM images. Methods: We developed and compared the performance of two machine learning models—DenseNet-201 CNN and XGBoost—trained on radiomic features with the ability to identify malignant tumors from benign ones. Feature importance was analyzed using SHAP and techniques of permutation importance that enhance both the global and case-specific interpretability of the models. Results: A few features that reflect tumor heterogeneity and morphology include GLCM Entropy, shape compactness, and surface-area-to-volume ratio. These performed excellently in diagnosis, with DenseNet-201 producing an accuracy of 92.4% and XGBoost at 89.7%. The analysis of feature interpretability ascertains its potential in early detection and boosting diagnostic confidence. Conclusions: The current work identifies the most important radiomic features and quantifies their diagnostic significance through a properly conducted feature selection process reflecting stability analysis. This provides the blueprint for feature-driven model interpretability in clinical applications. Radiomics features have great value in the automated diagnosis of lung cancer, especially when combined with machine learning models. This might improve early detection and open personalized diagnostic strategies for precision oncology. Full article

Lateral flow assays are widely used in point-of-care diagnostics but face challenges in sensitivity and accuracy when detecting low analyte concentrations, such as thyroid-stimulating hormone biomarkers. This study aims to enhance assay performance by leveraging textural features and hybrid artificial intelligence models. A modified Gray-Level Co-occurrence Matrix, termed the Averaged Horizontal Multiple Offsets Gray-Level Co-occurrence Matrix, was utilised to compute the textural features of the biosensor assay images. Significant textural features were selected for further analysis. A deep learning Convolutional Neural Network model was employed to extract features from these textural features. Both traditional machine learning models and hybrid artificial intelligence models, which combine Convolutional Neural Network features with traditional algorithms, were used to categorise these textural features based on the thyroid-stimulating hormone concentration levels. The proposed method achieved accuracy levels exceeding 95%. This pioneering study highlights the utility of textural aspects of assay images for accurate predictive disease modelling, offering promising advancements in diagnostics and management within biomedical research. Full article

Boundary contour determination during seedling image segmentation is critical for accurate object detection and morphological characterization in agricultural machine vision systems. The traditional manual annotation for segmentation is labor-intensive, time-consuming, and prone to errors, especially in controlled environments with complex backgrounds. These errors can affect the accuracy of detecting phenotypic traits, like shape, size, and width. To address these issues, this study introduced a method that integrated image features and a support vector machine (SVM) to improve boundary contour determination during segmentation, enabling real-time detection and monitoring. Seedling images (pepper, tomato, cucumber, and watermelon) were captured under various lighting conditions to enhance object–background differentiation. Histogram equalization and noise reduction filters (median and Gaussian) were applied to minimize the illumination effects. The peak signal-to-noise ratio (PSNR) and the structural similarity index measure (SSIM) were used to select the clip limit for histogram equalization. The images were analyzed across 18 different color spaces to extract the color features, and six texture features were derived using the gray-level co-occurrence matrix (GLCM) method. To reduce feature overlap, sequential feature selection (SFS) was applied, and the SVM was used for object segmentation. The SVM model achieved 73% segmentation accuracy without SFS and 98% with SFS. Segmentation accuracy for the different seedlings ranged from 81% to 98%, with a low boundary misclassification rate between 0.011 and 0.019. The correlation between the actual and segmented contour areas was strong, with an R² up to 0.9887. The segmented boundary contour files were converted into annotation files to train a YOLOv8 model, which achieved a precision ranging from 96% to 98.5% and a recall ranging from 96% to 98%. This approach enhanced the segmentation accuracy, reduced manual annotation, and improved the agricultural monitoring systems for plant health management. The future direction involves integrating this system with advanced methods to address overlapping image segmentation challenges, further enhancing the real-time seedling monitoring and optimizing crop management and productivity. Full article

Aboveground biomass (AGB) estimates derived from Landsat’s spectral bands are limited by spectral saturation when AGB densities exceed 150–300 Mg ${\mathrm{ha}}^{-1}$. Statistical features that characterize image texture have been proposed as a means to alleviate spectral saturation. However, apart from Gray Level Co-occurrence Matrix (GLCM) statistics, many spatial feature engineering techniques (e.g., morphological operations or edge detectors) have not been evaluated in the context of forest AGB estimation. Moreover, many prior investigations have been constrained by limited geographic domains and sample sizes. We utilize 176 lidar-derived AGB maps covering ∼9.3 million ha of forests in the Pacific Northwest of the United States to construct an expansive AGB modeling dataset that spans numerous biophysical gradients and contains AGB densities exceeding 1000 Mg ${\mathrm{ha}}^{-1}$. We conduct a large-scale inter-comparison of multiple spatial feature engineering techniques, including GLCMs, edge detectors, morphological operations, spatial buffers, neighborhood vectorization, and neighborhood similarity features. Our numerical experiments indicate that statistical features derived from GLCMs and spatial buffers yield the greatest improvement in AGB model performance out of the spatial feature engineering strategies considered. Including spatial features in Random Forest AGB models reduces the root mean squared error (RMSE) by 9.97 Mg ${\mathrm{ha}}^{-1}$. We contextualize this improvement model performance by comparing to AGB models developed with multi-temporal features derived from the LandTrendr and Continuous Change Detection and Classification algorithms. The inclusion of temporal features reduces the model RMSE by 18.41 Mg ${\mathrm{ha}}^{-1}$. When spatial and temporal features are both included in the model’s feature set, the RMSE decreases by 21.71 Mg ${\mathrm{ha}}^{-1}$. We conclude that spatial feature engineering strategies can yield nominal gains in model performance. However, this improvement came at the cost of increased model prediction bias. Full article

The objective of this study was to evaluate the effectiveness of utilizing radiomic features from radiation planning computed tomography (CT) scans in predicting tumor progression among patients with cervical cancers. A retrospective analysis was conducted on individuals who underwent radiotherapy for cervical cancer between 2015 and 2020, utilizing an institutional database. Radiomic features, encompassing first-order statistical, morphological, Gray-Level Co-Occurrence Matrix (GLCM), Gray-Level Run Length Matrix (GLRLM), and Gray-Level Dependence Matrix (GLDM) features, were extracted from the primary cervical tumor on the CT scans. The study encompassed 112 CT scans from patients with varying stages of cervical cancer ((FIGO Staging of Cervical Cancer 2018): 24% at stage I, 47% at stage II, 21% at stage III, and 10% at stage IV). Of these, 31% (n = 35/112) exhibited tumor progression. Univariate feature analysis identified three morphological features that displayed statistically significant differences (p < 0.05) between patients with and without progression. Combining these features enabled a classification model to be developed with a mean sensitivity, specificity, accuracy, and AUC of 76.1% (CI 1.5%), 70.4% (CI 4.1%), 73.6% (CI 2.1%), and 0.794 (CI 0.029), respectively, employing nested ten-fold cross-validation. This research highlights the potential of CT radiomic models in predicting post-radiotherapy tumor progression, offering a promising approach for tailoring personalized treatment decisions in cervical cancer. Full article

Background: This study aimed to analyze image biomarkers of the parotid glands in ultrasonography images with reduced shape effects, providing a reference for the radiomic diagnosis of parotid gland lesions. Methods: Ultrasound (US) and sialography images of the parotid glands, acquired from September 2019 to March 2024, were reviewed along with their clinical information. Parotid glands diagnosed as within the normal range were included. Overall, 91 US images depicting the largest portion of the parotid glands were selected for radiomic feature extraction. Regions of interest were drawn twice on 50 images using different shapes to assess the intraclass correlation coefficient (ICC). Feature dimensions were statistically reduced by selecting features with an ICC > 0.8 and applying four statistical algorithms. The selected features were used to distinguish age and sex using the four classification models. Classification performance was evaluated using the area under the receiver operating characteristic curve (AUC), recall, and precision. Results: The combinations of the information gain ratio algorithm or stochastic gradient descent and the naïve Bayes model showed the highest AUC for both age and sex classification (AUC = 1.000). The features contributing to these classifications included the first-order and gray-level co-occurrence matrix (high-order) features, particularly discretized intensity skewness and kurtosis, intensity skewness, and GLCM angular second moment. These features also contributed to achieving one of the highest recall (0.889) and precision (0.926) values. Conclusions: The two features were the most significant factors in discriminating radiomic variations related to age and sex in US images with reduced shape effects. These radiomic findings should be assessed when diagnosing parotid gland pathology versus normal using US images and radiomics in a heterogeneous population. Full article

Glaucoma, a primary cause of irreversible blindness, necessitates early detection to prevent significant vision loss. In the literature, fundus imaging is identified as a key tool in diagnosing glaucoma, which captures detailed retina images. However, the manual analysis of these images can be time-consuming and subjective. Thus, this paper presents an automated system for glaucoma detection using fundus images, combining diverse feature extraction methods with advanced classifiers, specifically Support Vector Machine (SVM) and AdaBoost. The pre-processing step incorporated image enhancement via Contrast-Limited Adaptive Histogram Equalization (CLAHE) to enhance image quality and feature extraction. This work investigated individual features such as the histogram of oriented gradients (HOG), local binary patterns (LBP), chip histogram features, and the gray-level co-occurrence matrix (GLCM), as well as their various combinations, including HOG + LBP + chip histogram + GLCM, HOG + LBP + chip histogram, and others. These features were utilized with SVM and Adaboost classifiers to improve classification performance. For validation, the ACRIMA dataset, a public fundus image collection comprising 369 glaucoma-affected and 309 normal images, was used in this work, with 80% of the data allocated for training and 20% for testing. The results of the proposed study show that different feature sets yielded varying accuracies with the SVM and Adaboost classifiers. For instance, the combination of LBP + chip histogram achieved the highest accuracy of 99.29% with Adaboost, while the same combination yielded a 65.25% accuracy with SVM. The individual feature LBP alone achieved 97.87% with Adaboost and 98.58% with SVM. Furthermore, the combination of GLCM + LBP provided a 98.58% accuracy with Adaboost and 97.87% with SVM. The results demonstrate that CLAHE and combined feature sets significantly enhance detection accuracy, providing a reliable tool for early and precise glaucoma diagnosis, thus facilitating timely intervention and improved patient outcomes. Full article

For regional desertification control and sustainable development, it is critical to quickly and accurately understand the distribution pattern and spatial and temporal changes of deserts. In this work, five different machine learning algorithms are used to classify different desert types on the Qinghai–Tibetan Plateau (QTP), and their classification performance is evaluated on the basis of their classification results and classification accuracy. Then, on the basis of the best classification model, the Shapely Additive Explanations (SHAP) method is used to clarify the contribution of each classification feature to the identification of desert types during the machine learning classification process, both globally and locally. Finally, the independent and interactive effects of each factor on desert change on the Qinghai-Tibetan Plateau during the study period are quantitatively analyzed via geodetector. The main results are as follows: (1) Compared with other classification algorithms (GTB, CART, KNN, and SVM), the RF classifier achieves the best performance in classifying QTP desert types, with an overall accuracy (OA) of 87.11% and a kappa coefficient of 0.83. (2) From the perspective of the overall classification of deserts, the five features, namely, elevation, slope, VV, VH, and GLCM, contribute most significantly to the features. In terms of the influence of each classification feature on the extraction of different types of deserts, the radar backscattering coefficient VV serves the most important role in distinguishing sandy deserts; the VH is helpful in distinguishing the four types of deserts: rocky desert, alpine cold desert, sandy deserts, and loamy desert; slope is more effective in distinguishing between the two desert types (rocky desert and alpine cold desert) and other types of deserts; and elevation has a significant role in the identification of alpine cold deserts; and the short-wave infrared band SR_B7 has an important role in the identification of salt crusts and saline deserts. (3) During the study period, the QTP deserts exhibited a reversing trend, and the proportion of desert area decreased from 28.62% to 26.20%. (4) Compared with other factors, slope, precipitation, elevation, vegetation type, and the human footprint have greater effects on changes in the QTP desert area, and the interactions among the factors affecting changes in the desert area all show bidirectional enhancement or nonlinear enhancement effects. Full article

Tree canopy cover is an important forest inventory parameter and a critical component for the in-depth mapping of forest fuels. This research examines the potential of employing single-date Sentinel-2 multispectral imagery, combined with contextual spatial information, to classify areas based on their tree cover density using Random Forest classifiers. Three spatial information extraction methods are investigated for their capacity to acutely detect canopy cover: two based on Gray-Level Co-Occurrence Matrix (GLCM) features and one based on segment statistics. The research was carried out in three different biomes in Greece, in a total study area of 23,644 km². Three tree cover classes were considered, namely, non-forest (cover < 15%), open forest (cover = 15%–70%), and closed forest (cover ≥ 70%), based on the requirements set for fuel mapping in Europe. Results indicate that the best approach identified delivers F1-scores ranging 70%–75% for all study areas, significantly improving results over the other alternatives. Overall, the synergistic use of spectral and spatial features derived from Sentinel-2 images highlights a promising approach for the generation of tree cover density information layers in Mediterranean regions, enabling the creation of additional information in support of the detailed mapping of forest fuels. Full article

Background: Alzheimer’s disease (AD) is a progressive and irreversible neurodegenerative condition that increasingly impairs cognitive functions and daily activities. Given the incurable nature of AD and its profound impact on the elderly, early diagnosis (at the mild cognitive impairment (MCI) stage) and intervention are crucial, focusing on delaying disease progression and improving patients’ quality of life. Methods: This work aimed to develop an automatic sMRI-based method to detect AD in three different stages, namely healthy controls (CN), mild cognitive impairment (MCI), and AD itself. For such a purpose, brain sMRI images from the ADNI database were pre-processed, and a set of 22 texture statistical features from the sMRI gray-level co-occurrence matrix (GLCM) were extracted from various slices within different anatomical planes. Different combinations of features and planes were used to feed classical machine learning (cML) algorithms to analyze their discrimination power between the groups. Results: The cML algorithms achieved the following classification accuracy: 85.2% for AD vs. CN, 98.5% for AD vs. MCI, 95.1% for CN vs. MCI, and 87.1% for all vs. all. Conclusions: For the pair AD vs. MCI, the proposed model outperformed state-of-the-art imaging source studies by 0.1% and non-imaging source studies by 4.6%. These results are particularly significant in the field of AD classification, opening the door to more efficient early diagnosis in real-world settings since MCI is considered a precursor to AD. Full article

Objectives: Subarachnoid Hemorrhage (SAH) is a serious neurological emergency case with a higher mortality rate. An automatic SAH detection is needed to expedite and improve identification, aiding timely and efficient treatment pathways. The existence of noisy and dissimilar anatomical structures in NCCT images, limited availability of labeled SAH data, and ineffective training causes the issues of irrelevant features, overfitting, and vanishing gradient issues that make SAH detection a challenging task. Methods: In this work, the water waves dynamic factor and wandering strategy-based Sand Cat Swarm Optimization, namely DWSCSO, are proposed to ensure optimum feature selection while a Parametric Rectified Linear Unit with a Stacked Convolutional Neural Network, referred to as PRSCNN, is developed for classifying grades of SAH. The DWSCSO and PRSCNN surpass current practices in SAH detection by improving feature selection and classification accuracy. DWSCSO is proposed to ensure optimum feature selection, avoiding local optima issues with higher exploration capacity and avoiding the issue of overfitting in classification. Firstly, in this work, a modified region-growing method was employed on the patient Non-Contrast Computed Tomography (NCCT) images to segment the regions affected by SAH. From the segmented regions, the wide range of patterns and irregularities, fine-grained textures and details, and complex and abstract features were extracted from pre-trained models like GoogleNet, Visual Geometry Group (VGG)-16, and ResNet50. Next, the PRSCNN was developed for classifying grades of SAH which helped to avoid the vanishing gradient issue. Results: The DWSCSO-PRSCNN obtained a maximum accuracy of 99.48%, which is significant compared with other models. The DWSCSO-PRSCNN provides an improved accuracy of 99.62% in CT dataset compared with the DL-ICH and GoogLeNet + (GLCM and LBP), ResNet-50 + (GLCM and LBP), and AlexNet + (GLCM and LBP), which confirms that DWSCSO-PRSCNN effectively reduces false positives and false negatives. Conclusions: the complexity of DWSCSO-PRSCNN was acceptable in this research, for while simpler approaches appeared preferable, they failed to address problems like overfitting and vanishing gradients. Accordingly, the DWSCSO for optimized feature selection and PRSCNN for robust classification were essential for handling these challenges and enhancing the detection in different clinical settings. Full article