Search Results (1,921)

Background/Objectives: The sella turcica and sphenoid sinus are anatomically adjacent structures within the cranial base and may reflect variations related to craniofacial development. However, evidence regarding their three-dimensional characteristics in individuals with impacted canines remains limited. This study aimed to evaluate the morphological, linear, and volumetric characteristics of the sella turcica and sphenoid sinus in individuals with unilateral palatally impacted maxillary canines using cone-beam computed tomography (CBCT). Methods: This study included CBCT scans of individuals with unilateral palatally impacted maxillary canines and a control group. Linear measurements and morphology of the sella turcica were assessed. Sella turcica volume was calculated using both a geometric formula and voxel-based three-dimensional segmentation. Sphenoid sinus pneumatization patterns and volumes were also evaluated. Agreement between volumetric measurement methods was assessed using Bland–Altman analysis, and correlations between sella turcica and sphenoid sinus volumes were also analyzed. Results: Most morphological and volumetric parameters of the sella turcica and sphenoid sinus were comparable between groups. Among the linear measurements, only sella width was significantly greater in the control group, whereas other dimensions showed no significant differences. The distribution of sella turcica morphology and sphenoid sinus pneumatization patterns was similar in both groups. No significant differences were observed in sella turcica or sphenoid sinus volumes. Bland–Altman analysis demonstrated good agreement between geometric and voxel-based volumetric measurements. In addition, no significant correlation was identified between sella turcica and sphenoid sinus volumes. Conclusions: Unilateral palatally impacted maxillary canines were not associated with substantial morphological or volumetric alterations of the sella turcica or sphenoid sinus. These findings suggest that variations in these cranial base structures have limited value as indicators of unilateral palatal canine impaction. Full article

Longan is an important economic fruit in tropical and subtropical regions, whose harvesting primarily relies on manual labor. Automated longan harvesting is key to improving the industry’s economic benefits but faces core challenges: mature pericarp is highly similar in color to fruiting mother branches, plus dense branches and severe leaf occlusion, leading to difficult cluster detection and fruiting branch segmentation. Herein, we propose a picking point localization method named PLYS-Longan integrating three customized core modules: Dynamic Convolution, Convolutional Gated Linear Unit (CGLU), and Dynamic Hyperbolic Tangent Activation (DYT) are introduced into YOLongan module to enhance the model’s ability to detect longan clusters. For SELongan module, Depthwise Over-parameterized Convolution (DO-Conv) and Ultra-light Subspace Attention (ULSA) are adopted to improve main branch segmentation precision. The PCLongan module then performs morphological erosion on the segmentation masks and calculates centroids to precisely determine the picking points. Experimental results show that the improved model achieves a mAP@50 of 90.1% (3.3% higher than baseline model) in object detection and a mIoU of 77.24% (1.75% improvement) in semantic segmentation, outperforming the various model significantly. This study provides an efficient and robust solution for longan picking point localization, laying a solid foundation for subsequent automated harvesting. Full article

Fibroblast growth factor 2 (FGF2) is expressed in retinal Müller glia cells, and its expression increases in response to photoreceptor degeneration. To investigate the physiological relevance of FGF2, we analyzed retinal morphology and cellular responses in Fgf2-deficient (Fgf2^−/−) mice. Loss of FGF2 did not affect photoreceptor survival, retinal vasculature, or retinal pigment epithelium (RPE) integrity. To further understand its role in retinal degeneration, Fgf2^−/− mice were crossed with Pde6b^STOP/STOP mice, a model of retinitis pigmentosa (RP). We then analyzed outer nuclear layer thickness, cone number, rod outer segments length, RPE morphology, and microglia number in Fgf2^−/− Pde6b^STOP/STOP and Pde6b^STOP/STOP mice. Although FGF2 was upregulated in degenerating photoreceptor cells in the Pde6b^STOP/STOP retina, its absence did not accelerate photoreceptor loss in Fgf2^−/− Pde6b^STOP/STOP mice. Interestingly, microglia numbers were significantly changed at early disease stages in Fgf2^−/− Pde6b^STOP/STOP retinas compared with Pde6b^STOP/STOP controls, suggesting that FGF2 modulates inflammatory signaling. Together, these results show that loss of FGF2 does not alter photoreceptor degeneration kinetics or retinal morphology, but may contribute to the regulation of early microglial accumulation during degeneration. Full article

The precise, real-time delineation of rice lodging areas constitutes a fundamental prerequisite for the adaptive operation of unmanned combine harvesters. However, existing deep learning methods struggle to resolve a critical limitation: achieving an optimal equilibrium between robust regional morphological perception—which is crucial for irregular lodging patterns—and the ultra-low computational overhead demanded by resource-constrained edge terminals. To address this specific constraint, StarNet-RiceSeg is proposed as a lightweight semantic segmentation network explicitly tailored for unmanned harvesters. Initially, the architecture incorporates the minimalist StarNet as its backbone. By leveraging the unique “Star Operation,” it implicitly maps features into a high-dimensional nonlinear space, thereby significantly augmenting feature discriminability while drastically curtailing computational overhead. Furthermore, to mitigate the misdetection issues stemming from the textural similarity between lodged and upright rice, the Rice Spatial Attention (RSA) module was designed. By intensifying feature interaction within the spatial dimension, this module steers the network to focus on the cohesive morphology of lodged regions while effectively suppressing background noise. Experiments conducted on a self-constructed high-resolution rice lodging dataset demonstrate that StarNet-RiceSeg achieves a mIoU of 94.42%, significantly outperforming mainstream models such as U-Net, DeepLabV3+, SegNet and HRNet. Notably, the model maintains a compact footprint with only 8.01 million parameters and a computational load as low as 9.32 GFLOPs. Following optimization with TensorRT, the system achieved a real-time inference speed of 32.51 FPS on the NVIDIA Jetson Xavier NX embedded platform. These results indicate that StarNet-RiceSeg provides a high-precision, low-latency solution for perceiving rice lodging areas in complex field environments, facilitating unmanned precision harvesting. Full article

Accurate rooftop segmentation from aerial imagery is essential for large-scale urban analysis, including applications such as solar potential assessment and urban monitoring. However, it remains constrained by the high cost of dense annotation and the limited generalisation of supervised models across heterogeneous urban morphologies. This study investigates binary rooftop segmentation for fine-tuning large image-editing foundation models using parameter-efficient Low-Rank Adaptation (LoRA). Using parts of Brisbane metropolitan dataset (split 80/20 into 97 training and 24 testing tiles), three paradigms were evaluated under a unified protocol: zero-shot image-editing models (including Gemini 3 Pro), a segmentation-first baseline (Segment Anything Model 3, SAM3), and LoRA-adapted diffusion models (FLUX.1 Kontext, FLUX.2, and Qwen Image Edit 2509) fine-tuned each 250 steps up to 5000 steps. Evaluated under zero-shot conditions, the generative models demonstrated varying levels of boundary fidelity. The Gemini model achieved a strong zero-shot baseline with [IoU, Dice] scores of [85%, 91%], followed by the SAM3 baseline, which also achieved a stable [84%, 91%] but exhibited increased false negatives in visually complex scenes. The tested diffusion models (FLUX.1 Kontext, FLUX.2, and Qwen) showed more limited initial spatial overlap, scoring [45%, 55%], [67%, 78%], and [33%, 46%], respectively. Following LoRA adaptation, the FLUX and Qwen models showed substantial improvements, with their respective [IoU, Dice] metrics increasing to [89%, 94%], [82%, 90%], and [87%, 93%]. FLUX.1 Kontext achieved the strongest overall performance at step 4250, yielding a mean IoU of 89% (SD = 3.16%) and a pixel accuracy exceeding 96%. These results demonstrate that parameter-efficient fine-tuning, combined with rigorous evaluation under class-imbalanced conditions, can transform general-purpose generative models into competitive, scalable spatial analysis tools that match or exceed both dedicated segmentation baselines and strong zero-shot multimodal models. Full article

Irreversible anterior segment blindness with preserved retinal integrity (e.g., dense corneal opacity) remains a major clinical challenge because effective sight-restoring options are limited. Here, we describe an intraocular micro-light-emitting diode (Micro-LED) epiretinal microdisplay intended to deliver patterned optical stimulation to intact photoreceptors by bypassing opaque anterior optics. The prototype was based on a color-capable VGA microdisplay (640 × 480 pixels) and operated at <30 mW under typical conditions. An ultra-thin flexible cable and a copper-mesh–reinforced polydimethylsiloxane (PDMS) encapsulation provided a compact, conformable intraocular package with high pixel density. We evaluated a monochromatic (green) prototype in a single beagle eye $(n=1)$ using a transscleral implantation approach and performed 7 days of postoperative follow-up with slit-lamp examination and multimodal imaging. Patterned stimulation via the implanted display elicited flash-evoked visual evoked potentials (VEPs) with consistent within-session waveform morphology, providing preliminary neurophysiological surrogate evidence of upstream visual pathway activation under the tested conditions in this single-animal pilot. The short-term postoperative course included transient hypotony and anterior segment inflammation, and implant rotation with associated inferior retinal detachment was observed by day 7, highlighting current biomechanical limitations. Beyond anterior segment opacity, the same intraocular optical interface could be explored as a modular light-delivery platform to pair with emerging retinal therapies (e.g., optogenetics), pending chronic safety and functional validation. This pilot large-animal study therefore provides a translationally relevant testbed while delineating key engineering constraints that must be addressed next. Full article

Background/Objectives: Spontaneous spinal subdural hematoma (SSSDH) is a rare and severe condition that causes rapid neurological decline. Spontaneous spinal epidural hematoma (SSEH) presents similarly but is more common, and surgical management differs because SSSDH requires an intradural approach. Few studies have assessed the reliability of magnetic resonance imaging (MRI) features used to distinguish SSSDH from SSEH in patients requiring surgery. Methods: We retrospectively reviewed 27 patients who underwent surgical evacuation of spinal hematomas at two institutions (2015–2025). Definitive hematoma location was determined intraoperatively. Four MRI features—shape (crescentic vs. biconvex), location (ventral vs. dorsal), craniocaudal length (<5 vs. ≥5 segments), and spinal region—were independently evaluated by two reviewers. Inter- and intra-rater reliability was assessed using agreement rate and Cohen’s kappa (κ) with 95% confidence intervals (95% CIs). Results: Among 27 cases, three (11.1%) were SSSDH and 24 were SSEH. Hematoma location, length, and spinal region demonstrated perfect inter- and intra-rater agreement (κ = 1.00). For hematoma shape, intra-rater agreement was good (96.2%, κ = 0.84; 95% CI 0.52–1.00), whereas inter-rater agreement was poor to fair (84.6%, κ = 0.26; 95% CI −0.25–0.77). Notably, two of the three SSSDHs demonstrated a biconvex configuration, and 83.3% of SSEHs also exhibited a biconvex morphology. Conclusions: MRI features such as hematoma location, extent, and spinal level were highly reproducible, whereas hematoma shape showed limited reliability. Although ventral hematomas most strongly suggest SSSDH, atypical SSEH presentations occur. When dorsal exposure reveals no epidural hematoma, intradural exploration should be promptly considered. Full article

Bone marrow cell classification and nucleus segmentation in microscopic images are fundamental tasks for computer-aided diagnosis of lymphocytic leukemia. However, bone marrow cells from different subtypes exhibit high morphological similarity, and structural information is often constrained under optical microscopic imaging, posing challenges for stable and effective feature representation. To address this issue, we propose MTLL (Multitask Model on Lymphocytic Leukemia), a novel multitask approach that performs cell classification and nucleus segmentation within a unified network to exploit their complementary information. The model constructs a hybrid backbone for shared feature representation based on a CNN-Transformer architecture, in which Fuse-MBConv modules are tightly integrated with multilayer multi-scale transformers to enable deep fusion of local texture and global semantic information. For the segmentation branch, we design an AM (Atrous Multilayer Perceptron) decoder that combines atrous spatial pyramid pooling with multilayer perceptrons to fuse multi-scale information and accurately delineate nucleus boundaries. The classification branch incorporates prior knowledge of cell nuclei structures to capture subtle variations in cellular morphology and texture, thereby enhancing the model’s ability to distinguish between leukemia subtypes. Experimental results demonstrate that the MTLL model significantly outperforms existing advanced single-task and multi-task models in both lymphocytic leukemia classification and cell nucleus segmentation. These results validate the effectiveness of the multi-task feature-sharing strategy for lymphocytic leukemia diagnosis using bone marrow microscopic images. Full article

Cervical cancer is one of the most prevalent and easily contracted diseases among women, significantly impacting their daily lives. Computer vision-based cervical cell morphology diagnosis technology can offer robust support for cervical cell analysis at a lower cost. However, the presence of a substantial number of overlapping cells in cervical images renders existing cell segmentation methods less accurate, thereby complicating the guidance of medical diagnosis. In this paper, we introduce a tristage Progressive Refinement method (PRefiner) for overlapping cell segmentation that decouples the traditional end-to-end pipeline, with the final stage specifically correcting anomalous results to enhance precision. We achieve separable overlapping cervical cell segmentation results through a cell nucleus locator, a single-cell segmenter, and a Segmentation Result Mask Refiner. Specifically, we employ a hybrid U-Net as the primary network for the cell nucleus locator and single-cell segmenter, which determines the position of the cell nucleus and procures the initial coarse segmentation result. In the mask refiner, we incorporate a conditional generation framework to address the perception decision problem and design a local–global dual-scale discriminator to ensure that the segmentation result aligns with the prior of a single-cell mask. Experimental results on CCEDD and ISBI2015 demonstrate that PRefiner achieves optimal performance by effectively resolving abnormal segmentations. Notably, our method improves the Dice coefficient of abnormal results from five different models by an average of 2.62% (ranging from 1.0% to 5.1%). Full article

Hypertension is a leading global health risk and requires accurate and continuous monitoring for effective management. Although photoplethysmography (PPG) is a promising non-invasive modality for cuffless blood pressure (BP) assessment, many existing approaches (especially raw-signal deep learning) are vulnerable to data leakage, overfitting on small datasets, limited interpretability, and poor performance on minority BP stages. To address these limitations, we propose a robust and physiologically grounded framework for multi-class BP stage classification based on interpretable PPG features. Our approach centers on a comprehensive multi-domain feature engineering pipeline that extracts 124 PPG features, including demographic, morphological, functional decomposition, spectral, nonlinear dynamics, and clinical composite indices. We apply rigorous preprocessing and feature selection prior to model training. We validate the framework on two datasets: PPG-BP dataset (657 segments, 4 classes) for benchmarking and PulseDB (283,773 segments, 3 classes) to assess scalability. We evaluate the proposed framework using a segment-level train/test split, appropriate for assessing intra-subject BP tracking after initial personalization. For the PulseDB dataset, this follows the protocol established by the dataset creators, while for the PPG-BP dataset, it enables direct comparison with prior work given practical dataset constraints. On PPG-BP, LightGBM trained on the selected features achieved macro-F1 = 0.78 and accuracy = 0.74, outperforming comparable deep-learning models. On the PulseDB, a custom Residual MLP achieved accuracy = 0.81 and macro-F1 = 0.79, supporting generalization at scale. These results show that the proposed feature-based approach can outperform complex end-to-end deep-learning models on small datasets while providing improved interpretability. This work establishes a reliable and transparent pathway toward clinically viable continuous BP staging, moving beyond black-box models toward physiologically grounded decision support. Ablation analysis reveals that engineered features provide most of the predictive power (F1 = 0.911), while raw PPG features alone achieve modest performance (F1 = 0.384). For the minority hypertension stage 2 (HT-2) class, a bootstrap 95% confidence interval of [0.762, 1.000] is reported, reflecting uncertainty due to limited sample size. Full article

Low-accessible Cultural Heritage, including hypogeal sites, rupestrian architectures, and fragile structures, represents a major challenge for conservation, documentation, and continuous monitoring. These limitations stem from multiple inaccessibility factors, classified as physical (morphological complexity), asset risk (microclimatic instability), health and safety (structural vulnerability), managerial (lack of public access), and cognitive (lack of documentation). This research aims to transform digital models from mere representational tools into integrated cognitive and operational systems supporting decision-making and preventive conservation. The proposed methodological workflow is structured into five main phases: Preliminary Knowledge and Multidisciplinary Data Structuring (Ph1. PK–MDS), Comprehensive Digital Survey (Ph2. CDS), Development of Integrated Digital Models (Ph3. IDMs), Advanced Diagnosis and Monitoring (Ph4. ADM) and the implementation of an Integrated Digital Environment for Hypogeal Heritage Management (Ph5. IDE). Ph4 operates on two complementary scales: at the site scale, range-based point clouds enable the semi-automatic identification of extensive decay patterns, such as biological colonization. At the detail scale, the Random Forest algorithm enables the segmentation and quantification of material loss on frescoed surfaces through a diachronic comparison of historical and current data. Validated on the San Pellegrino complex in Matera, selected as a paradigmatic case study of low-accessibility hypogeal sites, representative of a broader system comprising approximately 150 rupestrian cult architectures, the methodology demonstrates how immersive digital environments function as shared knowledge spaces, supporting more informed, inclusive, and resilient heritage conservative management. Full article

For projects such as tailings ponds, slopes, and foundations, loose materials such as rock, slag, and sand, which are composed of particles, often have low cohesion and rely mainly on friction to maintain stability. The shear strength parameters, namely, the internal friction angle and cohesion, are the core parameters that describe the mechanical properties of materials and are directly related to the engineering stability of the above projects. The shear strength properties of loose media are related to the geometric morphological characteristics of particles. Particles with high irregularity will increase the bite and friction of the contact interface between particles, thereby affecting the overall peak shear strength of the material. This study takes sand as the research object. Based on the Mask R-CNN algorithm in deep learning, a sand particle image dataset consisting of single, contact, and sand surface particles is established. An image segmentation model that can identify particles on the surface of the sand layer and obtain the corresponding particle mask is trained; a Python 3.11.4 program is written to automatically calculate seven characteristic parameters of particle morphological characteristics parameters, including the Feret major diameter, the particle Feret minor diameter, the particle aspect ratio, the particle roundness, the comprehensive shape coefficient, the roughness, and the convexity through the particle mask. This method can obtain the overall morphological characteristics of sand particles in real time and is a particle processing method that is a prerequisite for the subsequent rapid prediction of the strength properties of granular materials. Full article

Laser Powder Bed Fusion for metals (PBF-LB/M) is a complex additive manufacturing process in which metal powder is selectively melted layer-by-layer to fabricate 3D parts. Process parameters critically influence the resulting microstructure in nickel alloys, with features such as melt pool marks, grain size and orientation, porosity, and cracks serving as key process signatures. These features are typically analyzed post-process to identify suboptimal conditions. This research aims to develop automated post-process measurement and analysis techniques using image processing, pattern recognition, and statistical learning to correlate process parameters with part quality. Optical microscopy images of build surfaces are analyzed using machine learning algorithms to evaluate porosity, grain size, and relative density in fabricated test coupons. Effect plots are generated to identify trends related to increasing energy density. A novel deep learning approach based on Mask R-CNN is used to detect and segment melt pool regions in optical microscopy images. From the segmented regions, melt pool dimensions—such as width, depth, and area—are extracted using bounding geometry coordinates. Manually labeled images (Type I and Type II) are used to train the model. A comparison between ResNet-50 and ResNet-101 backbones shows that the ResNet-50-based model (Model 2) achieves superior performance, with lower training loss (0.1781 vs. 0.1907) and validation loss (8.6140 vs. 9.4228). Quantitative evaluation using the Jaccard index, precision, and recall metrics shows that the ResNet-101 backbone outperforms ResNet-50, achieving about 4% higher mean Intersection-over-Union, with values of 0.85 for Type I and 0.82 for Type II melt pools, where Type I is detected more accurately due to its more regular morphology and clearer boundaries. By extending Faster R-CNNs with a mask prediction branch, the method allows for precise melt pool measurements, providing valuable insights into process quality and dimensional accuracy, and aiding in the detection of defects in PBF-LB-fabricated parts. Full article

Chronic kidney disease (CKD) is a major global health burden leading to a loss of kidney function via podocyte damage, a non-regenerative renal cell type. Early detection of podocyte injury is crucial but remains limited, highlighting the need for non-invasive biomarkers. Therefore, we analysed urinary exosomal microRNAs (miRNAs) in relation to podocyte morphology in biopsies from 65 CKD patients, including focal segmental glomerulosclerosis (FSGS), minimal change disease (MCD) and healthy controls. Global profiling distinguished CKD patients from controls, with miR-606 consistently upregulated and miR-431 downregulated. In podocytopathies, MCD displayed a predominantly suppressed miRNA profile, with miR-141, miR-429, and miR-660 as key candidates, whereas FSGS exhibited elevated miR-181c, miR-3610, miR-663b, miR-4651, and miR-429. Super-resolution morphometry revealed diffuse foot process effacement in MCD and heterogeneous, focally disrupted architecture in FSGS, providing a structural context for the molecular findings. Regression analyses linked these miRNAs to filtration slit density and length, proteinuria, and 25-Hydroxy-vitamin-D3 levels, integrating molecular, structural, and clinical readouts. These results define a coherent miRNA signature of podocyte injury that distinguishes CKD entities and correlates molecular changes with disease severity. Combining urinary exosomal miRNAs with morphometric analysis facilitates early, non-invasive identification of podocyte damage, enabling earlier therapeutic intervention in podocytopathies. Full article

Homopolymer addition is a widely used strategy to dilate the microdomain spacing of block copolymers, yet the attainable dilation is often limited by macrophase separation in conventional blends at elevated homopolymer loading. In this work, we investigate an architectural route to suppress macrophase separation while retaining homopolymer-driven dilation: a covalently hybridized bottlebrush copolymer (CH-BBC), a copolymacromer-like bottlebrush architecture in which symmetric AB diblock side chains and A-type homopolymer side chains are covalently grafted to a common backbone. Using dissipative particle dynamics (DPD) simulations, we directly compare the phase behavior of CH-BBC melts with that of composition-matched blends of symmetric AB diblocks and A-type homopolymers. Across the explored window, CH-BBC exhibits microphase morphologies and disorder without an observable two-phase region, whereas the corresponding blends show extensive two-phase coexistence at elevated homopolymer loading. Lamellar analysis and one-dimensional density decompositions further reveal that CH-BBC enables substantially larger microphase dilation and stronger selective swelling of the A-rich domain because tethered A-type homopolymer segments preferentially occupy and dilate the A-rich domain interior while diblock A segments remain localized near interfaces. Full article