Search Results (3,239)

Reliable and efficient detection of persimmons provides the foundation for precise maturity evaluation. Persimmon ripeness detection remains challenging due to small target sizes, frequent occlusion by foliage, and motion- or focus-induced blur that degrades edge information. This study proposes YOLOv11-4ConvNeXtV2, an enhanced detection framework that integrates a ConvNeXtV2 backbone with Fully Convolutional Masked Auto-Encoder (FCMAE) pretraining, Global Response Normalization (GRN), and Single-Head Self-Attention (SHSA) mechanisms. We present a comprehensive persimmon dataset featuring sub-block segmentation that preserves local structural integrity while expanding dataset diversity. The model was trained on 4921 annotated images (original 703 + 6 × 703 augmented) collected under diverse orchard conditions and optimized for 300 epochs using the Adam optimizer with early stopping. Comprehensive experiments demonstrate that YOLOv11-4ConvNeXtV2 achieves 95.9% precision and 83.7% recall, with mAP@0.5 of 88.4% and mAP@0.5:0.95 of 74.8%, outperforming state-of-the-art YOLO variants (YOLOv5n, YOLOv8n, YOLOv9t, YOLOv10n, YOLOv11n, YOLOv12n) by 3.8–6.3 percentage points in mAP@0.5:0.95. The model demonstrates superior robustness to blur, occlusion, and varying illumination conditions, making it suitable for deployment in challenging maturity detection environments. Full article

Background: Renal Cell Carcinoma (RCC) is the most common type of kidney cancer and requires accurate histopathological grading for effective prognosis and treatment planning. However, manual grading is time-consuming, subjective, and susceptible to inter-observer variability. Objective: This study proposes EAT-Net (Efficient Attention Transformer Network), a dual-stream deep learning model designed to automate and enhance RCC grade classification from histopathological images. Method: EAT-Net integrates EfficientNetB0 for local feature extraction and a Vision Transformer (ViT) stream for capturing global contextual dependencies. The architecture incorporates Squeeze-and-Excitation (SE) modules to recalibrate feature maps, improving focus on informative regions. The model was trained and evaluated on two publicly available datasets, KMC-RENAL and RCCG-Net. Standard preprocessing was applied, and the model’s performance was assessed using accuracy, precision, recall, and F1-score. Results: EAT-Net achieved superior results compared to state-of-the-art models, with an accuracy of 92.25%, precision of 92.15%, recall of 92.12%, and F1-score of 92.25%. Ablation studies demonstrated the complementary value of the EfficientNet and ViT streams. Additionally, Grad-CAM visualizations confirmed that the model focuses on diagnostically relevant areas, supporting its interpretability and clinical relevance. Conclusion: EAT-Net offers an accurate, and explainable framework for RCC grading. Its lightweight architecture and high performance make it well-suited for clinical deployment in digital pathology workflows. Full article

Background/Objectives: Glioblastoma (GBM) is an aggressive primary brain tumor marked by diffuse infiltration into surrounding brain tissue. The peritumoral zone often appears normal on imaging yet harbors microscopic invasion. While perfusion-based studies, such as arterial spin labeling (ASL), have profiled this region, longitudinal radiomic monitoring remains limited. This study investigates delta radiomics using multiparametric MRI (mpMRI) in a GBM mouse model to track subtle peritumoral changes over time. Methods: A G7 GBM xenograft model was established in nine nude mice, imaged at 9- and 12 weeks post-implantation using MRI (T1W, T2W, T2 mapping, DWI-ADC, FA, and ASL) and co-registered histopathology (H&E, HLA staining). Tumor and peritumoral regions were manually segmented, and 107 radiomic features (shape, first-order, texture) were extracted per sequence and histology. The delta features were calculated and compared between timepoints. Results: The robust T2W texture and T2 map first-order features demonstrated the greatest sensitivity and reproducibility in capturing temporal peritumoral brain zone changes, distinguishing between time points used by K-mean. Conclusions: Delta radiomics offers added value over static analysis for early monitoring of peritumoral brain zone changes. The first-order and texture features of radiomics could serve as robust biomarkers of peritumoral invasion. These findings highlight the potential of longitudinal MRI-based radiomics to characterize glioblastoma progression and inform translational research. Full article

Background/Objectives: Satellite DNAs (satDNAs) are tandemly repeated sequences that play essential roles in chromosome structure, genome organization, and evolution. Despite their importance, the satellitome (the complete collection of satDNAs) of most avian lineages remains unexplored. We sought to describe the repeatome of three trogonid species, Trogon surrucura, T. melanurus, and Apaloderma vittatum with a focus on the satellitome to evaluate the general features of this lineage. Methods: Herein, we provide the first comparative characterization of the repeatome, with a particular focus on the comparative characterization of satDNAs in three trogonid species: T. surrucura, T. melanurus, and A. vittatum. Using a combination of bioinformatic pipelines and cytogenetic approaches. Results: We identified 16 satDNA families in T. surrucura, 15 in T. melanurus, and only 3 in A. vittatum. Sequence comparisons revealed that five families are shared between the two Trogon species, consistent with the library hypothesis, whereas no satDNAs were shared with A. vittatum. While both Trogon species exhibited a predominance of GC-rich repeats, A. vittatum represents the first bird described with a satellitome dominated by AT-rich satDNAs. In situ mapping in T. surrucura revealed chromosome-specific satDNAs restricted to pairs 1 and 2 and a Z-specific repeat that was strongly accumulated on its long arms, an atypical feature among birds. Conversely, the W chromosome showed a surprisingly low number of satDNAs, limited to centromeric signals. Conclusions: Our results reveal highly divergent satellitome landscapes among trogonids, characterized by lineage-specific differences in repeat composition, abundance, and chromosomal distribution. These findings support the view that satDNAs are dynamic genomic elements, whose amplification, loss, and chromosomal redistribution can influence genome architecture and play a role in avian speciation. Full article

Weeds continue to pose a serious challenge to agriculture, reducing both the productivity and quality of crops. In this paper, we explore how modern deep learning, specifically Vision Transformers (ViTs), can help address this issue through fast and accurate weed classification. We developed a transformer-based model trained on the DeepWeeds dataset, which contains images of nine different weed species collected under various environmental conditions, such as changes in lighting and weather. By leveraging the ViT architecture, the model is able to capture complex patterns and spatial details in high-resolution images, leading to improved prediction accuracy. We also examined the effects of model optimization techniques, including fine-tuning and the use of pre-trained weights, along with different strategies for handling class imbalance. While traditional oversampling actually hurt performance, dropping accuracy to 94%, using class weights alongside strong data augmentation boosted accuracy to 96.9%. Overall, our ViT model outperformed standard Convolutional Neural Networks, achieving 96.9% accuracy on the held-out test set. Attention-based saliency maps were inspected to confirm that predictions were driven by weed regions, and model consistency under location shift and capture perturbations was assessed using the diverse acquisition sites in DeepWeeds. These findings show that with the right combination of model architecture and training strategies, Vision Transformers can offer a powerful solution for smarter weed detection and more efficient farming practices. Full article

Under the tightening carbon reduction policies, port microgrids face the challenge of optimizing the installed capacity of multiple power generation types to reduce operating costs and increase renewable energy penetration. We develop a bi-level cost-optimization framework in which the upper level decides long-term capacities (PV, wind, gas turbine, bio-fuel unit, and battery energy storage), and the lower level dispatches a multi-energy port microgrid (electricity–heat–cold) on an hourly basis with frequency regulation services. To ensure rigor and reproducibility, we (i) move the methodology upfront and formalize all constraints, (ii) provide a dedicated data–preprocessing pipeline for multi-region 50/60 Hz frequency time series, and (iii) map a policy intensity index to a carbon price and/or an annual cap used in the objective/constraints. The bi-level MILP is solved by a column-and-constraint generation algorithm with optimality gap control. We report quantitative metrics—annualized total cost, CO₂ emissions (t), renewable shares (%), and regulation cycles—across scenarios. Results show consistent cost–carbon trade-offs and robust capacity shifts toward storage and biofuel as policy tightens. All inputs and scripts are organized for exact replication. Full article

More than 60% of earth excavation operations have been accomplished by various excavators. However, complex working loads always cause the fracture failure of excavator sticks because of insufficient strength. For prolonging the service life of excavator stick, a structural optimization design method based on the improved mayfly optimization algorithm (TTL-MA) is proposed to improve the stiffness of excavator stick. Firstly, by using the central composite design (CCD) method, 161 sets of simulation samples are obtained with eight selected structural design parameters of excavator stick. Then, relying on the simulation samples, an agent model between the excavator stick’s structural design parameters and the structural quality objectives, deformation, first-order minimum intrinsic frequency, and stress is constructed by using a Backpropagation neural network (BPNN). Finally, to further enhance the optimization search capability of the Mayfly Algorithm (MA), three improvement strategies were incorporated: Tent chaotic mapping for mayfly population initialization, adaptive t-distribution perturbation for velocity updating, and Lévy flight strategy for enhanced position updating. The results show that under the three constraints of the maximum equivalent von Mises stress σ_max ≤ 150 MPa, maximum deformation δ_max ≤ 2.5 mm, and the first-order minimum intrinsic frequency H_min ≥ 55 Hz, the optimized excavator stick reduces the mass and maximum stress by 7.9% and 11.9%, respectively. The improved mayfly optimization algorithm has strong optimization ability for the optimization design of excavator stick structure, which can provide a reference for similar complex engineering machinery structure optimization problems. Full article

Unmanned aerial vehicle (UAV) images have increasingly become important data for accurate mapping of tidal marsh vegetation. They are particularly important for determining what spatial resolution is needed because UAV imaging requires a trade-off between spatial resolution and imaging extent. However, there are still insufficient studies for assessing the effects of spatial resolution on the classification accuracy of tidal marsh vegetation. This study utilized UAV images with spatial resolutions of 2 cm, 5 cm, and 10 cm, respectively, to classify seven tidal marsh plots with different vegetation complexities in the Yellow River Delta (YRD), China, using the object-oriented example-based feature extraction with support vector machine approach and the pixel-based random forest classifier, and compared the differences in vegetation classification accuracy. This study indicated the following: (1) Vegetation classification varied at different spatial resolutions, with a difference of 0.95–8.76% between the highest and lowest classification accuracy for different plots. (2) Vegetation complexity influenced classification accuracy. Classification accuracy was lower when the relative dominance and proportional abundance of P. australis and T. chinensis were higher in the plots. (3) Considering the trade-off between classification accuracy and imaging extent, UAV data with 5 cm spatial resolution were recommended for tidal marsh vegetation classification in the YRD or similar vegetation complexity regions. Full article

Background/Objectives: Magnetisation Transfer (MT) MRI is used for neuro-degenerative disorders, including Multiple Sclerosis (MS), providing an indirect measure of large biomolecular MR signal sources which cannot be observed directly because their typical T2 is usually much shorter than the echo time (TE) of conventional MR sequences. We investigated a 3D-radial Zero Time of Echo (ZTE) MT-weighted sequence with potentially enhanced sensitivity to short-T2 MR signals indirectly (via MT weighting) and directly (due to the short TE). Methods: The sequence runs on a human 7T MR scanner, producing whole-brain MT-weighted images with isotropic 0.8 mm resolution in 6.5 minutes. One RF pulse is used to suppress the fat signal and generate MT weighting, reducing RF power deposition to moderate levels. The small excitation pulses and the “quasi-adiabatic” MT pulse mitigate the negative effects of inhomogeneous transmit RF fields observed at 7T in the human head, facilitating the generation of uniform Magnetisation Transfer Ratio (MTR) maps. Results: Results from a biologic phantom, a healthy volunteer, and an MS patient illustrate important imaging features of the “SilentMT” sequence. When the MS patient images were compared with Fluid Attenuated Inversion Recovery (FLAIR) images taken on the same patient at 1.5T and 7T, SilentMT was able to detect all the MS lesions observed on the “reference truth” 1.5T FLAIR; 7T FLAIR, however, failed to detect some lesions in the temporal lobe and brain stem. SilentMT detected a lesion which was not immediately apparent on either FLAIR image. Increased MTR was observed in some regions of the brain of the MS patient, notably the left temporal lobe. Conclusions: This initial investigation of an MT-weighted ZTE sequence shows evidence that it may be more sensitive to pathology in a patient with MS. Full article

Background: Pediatric acute myeloid leukemia (pAML) is the second most common type of childhood leukemia, behind acute lymphoblastic leukemia. High-throughput technologies have enabled the identification of increasing molecular alterations linked to AML prognosis, revealing genomic heterogeneity among individual patients and providing clinically valuable diagnostic and prognostic information. This study systematically analyzed the correlation between high-frequency mutated genes and prognosis in pAML by performing whole-transcriptome sequencing (WTS) of bone marrow samples from newly diagnosed AML children in Southwest China and mapping their genetic profiles. Methods: pAML patients treated at the Department of Hematology and Oncology, Children’s Hospital of Chongqing Medical University, from January 2015 to October 2024, were enrolled, and WTS was performed. The study described the frequency, pathogenicity classification, and risk stratification of mutation genes and fusion genes, and constructed a genetic landscape. For high-frequency pAML mutations, the impact on early induction remission rate (CR) and long-term event-free survival (EFS) was evaluated. Results: A total of 134 pediatric AML patients from Southwest China were included, with a male-to-female ratio of 74:60 and a median diagnosis age of 5.96 years. Based on pathogenicity classification using WTS, fusion genes were categorized into level 1, level 2, and level 3 genes, as well as mutation genes. The study identified five fusion genes of level 1, the most frequent being RUNX1::RUNX1T1 (32/134, 23.88%), KMT2A rearrangements (29/134, 21.64%), and CBFB::MYH11 (13/134, 9.7%). Sixteen mutation genes of level 1 were detected, seven of which recurred in over 5% of patients, including NRAS (31/134, 23.13%), FLT3 (25/134, 18.66%), KIT (24/134, 17.91%), CEBPA (14/134, 10.45%), WT1 (13/134, 9.7%), KRAS (11/134, 8.2%), and PTPN11 (7/134, 5.22%). Sex-based analysis revealed that PTPN11 mutations were significantly more frequent in males (9.45% vs. 0%, p = 0.023), as were KIT mutations (24.32% vs. 10.00%, p = 0.044). Risk-stratified analysis showed that WT1 mutations (14.13% vs. 0%, p = 0.031) and FLT3-ITD mutations (13.19% vs. 0%, p = 0.042) were enriched in intermediate- and high-risk groups, whereas CEBPA (25.64% vs. 5.43%, p = 0.012), KIT (35.90% vs. 10.87%, p = 0.003), and KIT-E8 (20.51% vs. 1.10%, p < 0.001) mutations were more prevalent in low-risk groups. Prognostic analysis indicated that PTPN11 and KIT mutations did not affect CR or EFS across sexes, nor did WT1, CEBPA, or KIT mutations influence outcomes by risk stratification. However, FLT3-ITD-positive patients had significantly lower CRs (χ² value = 11.965, p = 0.007), although EFS differences were nonsignificant. In contrast, WT1 mutations were associated with inferior EFS compared to wild-type (p = 0.036). Furthermore, the univariate and multivariate Cox regression revealed consistent results with the above findings, indicating that WT1 mutation was an independent adverse prognostic factor for EFS (HR = 2.400, 95% CI: 1.101–5.233, p = 0.028). The results of univariate and multivariate logistic regression analyses also confirmed that FLT3-ITD mutation was an independent predictor of initial treatment response in our cohort (OR = 10.699, 95% CI: 2.108–54.302, p = 0.004). Conclusions: This study delineated the genetic landscape of pAML in Southwest China and explored the prognostic value of gene fusions and mutations in early and long-term outcomes. These findings provide a foundation for understanding the genetic heterogeneity of pAML and offer evidence for the development of precision medicine approaches. Full article

Background: Pneumothorax (PTX) requires rapid recognition in emergency and critical care. Lung ultrasound (LUS) offers a fast, radiation-free alternative to computed tomography (CT), but its accuracy is limited by operator dependence. Artificial intelligence (AI) may standardize interpretation and improve performance. Methods: This retrospective single-center study included 46 patients (23 with CT-confirmed PTX and 23 controls). Sixty B-mode and M-mode frames per patient were extracted using a Clarius C3 HD3 wireless device, yielding 2760 images. CT served as the diagnostic reference. Experimental studies were conducted within the framework of three scenarios. Transformer-based models, Vision Transformer (ViT) and DINOv2, were trained and tested under two scenarios: random frame split and patient-level split. Also, Random Forest (RF) and eXtreme Gradient Boosting (XGBoost) classifiers were trained on the feature maps extracted by using Video Vision Transformer (ViViT) for ultrasound video sequences in Scenario 3. Model performance was evaluated using accuracy, sensitivity, specificity, F1-score, and area under the ROC curve (AUC). Results: Both transformers achieved high diagnostic accuracy, with B-mode images outperforming M-mode inputs in the first two scenarios. In Scenario 1, ViT reached 99.1% accuracy, while DINOv2 achieved 97.3%. In Scenario 2, which avoided data leakage, DINOv2 performed best in the B-mode region (90% accuracy, 80% sensitivity, 100% specificity, F1-score 88.9%). ROC analysis confirmed strong discriminative ability, with AUC values of 0.973 for DINOv2 and 0.964 for ViT on B-mode images. Also, both RF and XGBoost classifiers trained on the ViViT feature maps reached 90% accuracy on the video sequences. Conclusions: AI-assisted LUS substantially improves PTX detection, with transformers—particularly DINOv2—achieving near-expert accuracy. Larger multicenter datasets are required for validation and clinical integration. Full article

Accurately predicting survival within patients diagnosed with diffuse glioma remains one of the most difficult issues in neuro-oncology. While most prior research has focused on multimodal fusion or clinical data, we introduce a modality-specific deep learning framework that employs preoperative MRI only to predict mortality outcomes using patient MRI scans. Using the UCSF-PDGM dataset containing structural, diffusion, and perfusion imaging of 495 glioma patients, we trained VGG16 models on every MRI modality individually, including T1, T2, FLAIR, SWI, DWI, ASL, HARDI-derived metrics, and segmentation maps. Our findings revealed that segmentation-based and diffusion-derived features, particularly FA or tensor eigenvalues, possessed the greatest predictive strength, surpassing those obtained from standard structural MRI in binary survival classifications. This approach of modality-specific model training allows for clearer explanations of the prediction process compared to fused approaches and is more practical in scenarios where not all types of MRI are performed on patients. This approach demonstrates the strong predictive power of individual MRI sequences for mortality in glioma cases, providing a modular, adaptable, and clinically actionable deep-learning framework. Additional enhancements can incorporate volumetric models, longitudinal imaging, and non-imaging datasets, including genomic and clinical information. Full article

This study presents an integrated framework for spatiotemporal mapping of carbon stocks in rubber plantations in Rayong Province, Eastern Thailand—an area undergoing rapid agricultural transformation and rubber expansion. Unlike most existing assessments that rely on Tier 1 IPCC defaults or coarse plantation age classes, our framework combines annual plantation age derived from Landsat time series, age-specific allometric growth models, and Tier 2 soil organic carbon (SOC) accounting. This enables fine-scale, age- and site-sensitive estimation of both tree and soil carbon. Results show that tree biomass dominates the carbon pool, with mean tree carbon stocks of 66.94 ± 13.1% t C ha⁻¹, broadly consistent with national field studies. SOC stocks averaged 45.20 ± 0.043% t C ha⁻¹, but were overwhelmingly inherited from pre-conversion land use (43.7 ± 0.042% t C ha⁻¹). Modeled SOC changes ($\mathsf{\Delta }\mathrm{SOC}$) were modest, with small gains (2.06 t C ha⁻¹) and localized losses (−9.96 t C ha⁻¹), producing a net mean increase of only 1.44 t C ha⁻¹. These values are substantially lower than field-based estimates (5–15 t C ha⁻¹), reflecting structural limitations of the global empirical ΔSOC model and reliance on generalized default parameters. Uncertainties also arise from allometric assumptions, generalized soil factors, and Landsat resolution constraints in smallholder landscapes. Beyond carbon, ecological trade-offs of rubber expansion—including biodiversity loss, soil fertility decline, and hydrological impacts—must be considered. By integrating methodological innovation with explicit acknowledgment of uncertainties, this framework provides a conservative but policy-relevant basis for carbon accounting, subnational GHG reporting, and sustainable land-use planning in tropical agroecosystems. Full article

Background/Objectives: Prostate cancer (PCa) has the highest incidence and mortality rates among men in Puerto Rico. However, screening and early detection programs remain limited and fragmented. This study presents the design and implementation of a community-based educational program to increase PCa screening and knowledge in three southeastern rural communities with high African ancestry and elevated PCa mortality. Methods: Conducted between 2021 and 2025, this mixed-method study followed Community Engagement principles and was guided by the Intervention Mapping framework. A Community Advisory Committee informed each step of the intervention, which included PSA and digital rectal examination (DRE) testing via a mobile clinic staffed by urologists. Pre- and post-tests measured knowledge gains and willingness to screen, while satisfaction surveys evaluated the program’s impact. Results: After the intervention, knowledge scores increased significantly (t = −5.5, p < 0.001), and 76% of participants reported greater confidence in making health decisions. In total, 95 men accessed screening services through a mobile clinic, 33 were referred for follow-up, and 4 PCa cases were detected. Conclusions: Combining culturally tailored education with accessible screening helped overcome sociocultural and structural barriers, showing promise for reducing PCa disparities in underserved Puerto Rican populations. Full article

Modern maize breeding worldwide relies on a broad range of molecular genetics research techniques. These technologies allow us to identify genomic regions associated with various phenotypic traits, including resistance to fungi of the genus Fusarium. Therefore, the aim of this publication was to identify new molecular markers linked to candidate genes that confer maize resistance to Fusarium fungi, using next-generation sequencing, association mapping, and physical mapping. In the study, a total of 5714 significant molecular markers related to maize plant resistance to Fusarium fungi were identified. Of these, 10 markers were selected that were significantly associated (with the highest LOD values) with the disease. These markers were identified on chromosomes 5, 6, 7, 8, and 9. The authors were particularly interested in two markers: SNP 4583014 and SilicoDArT 4579116. The SNP marker is located on chromosome 5, in exon 8 of the gene encoding alpha-mannosidase I MNS5. The SilicoDArT marker is located 240 bp from the gene for peroxisomal carrier protein on chromosome 8. Our own research and the presented literature review indicate that both these genes may be involved in biochemical reactions triggered by the stress caused by plant infection with Fusarium fungal spores. Molecular analyses indicated their role in resistance processes, as resistant varieties responded with an increase in the expression level of these genes at various time points after plant inoculation with Fusarium fungal spores. In the negative control, which was susceptible to Fusarium, no significant fluctuations in the expression levels of either gene were observed. Analyses concerning the identification of Fusarium fungi showed that the most abundant fungi on the infected maize kernels were Fusarium poae and Fusarium culmorum. Individual samples were very sparsely colonized by Fusarium or not at all. By using various molecular technologies, we identified genomic regions associated with maize resistance to Fusarium fungi, which is of fundamental importance for understanding these regions and potentially manipulating them. Full article