Search Results (56)

Background/Objectives: Accurate identification of deciduous and permanent teeth on panoramic radiographs (PRs) during mixed dentition is fundamental for early detection of eruption disturbances, yet relies heavily on clinician experience due to developmental variability. This study aimed to develop a deep learning model for automated tooth detection and segmentation in pediatric PRs during mixed dentition. Methods: A retrospective dataset of 250 panoramic radiographs from patients aged 6–13 years was analyzed. A customized YOLOv11-based model was developed using a novel hybrid pre-annotation strategy leveraging transfer learning from 650 publicly available adult radiographs, followed by expert manual refinement. Performance evaluation utilized mean average precision (mAP), F1-score, precision, and recall metrics. Results: The model demonstrated robust performance with mAP_0.5 = 0.963 [95%CI: 0.944–0.983] and macro-averaged F1-score = 0.953 [95%CI: 0.922–0.965] for detection. Segmentation achieved mAP_0.5 = 0.890 [95%CI: 0.857–0.923]. Stratified analysis revealed excellent performance for permanent teeth (F1 = 0.977) and clinically acceptable accuracy for deciduous teeth (F1 = 0.884). Conclusions: The automated system achieved near-expert accuracy in detecting and segmenting teeth during mixed dentition using an innovative transfer learning approach. This framework establishes reliable infrastructure for AI-assisted diagnostic applications targeting eruption or developmental anomalies, potentially facilitating earlier detection while reducing clinician-dependent variability in mixed dentition evaluation. Full article

Artificial intelligence (AI) is an important tool for improving agricultural tasks. In particular, object detection methods based on convolutional neural networks (CNNs) enable the detection and classification of objects directly in the field. Combined with unmanned aerial vehicles (UAVs, drones), these methods allow efficient crop monitoring. The primary challenge is to develop models that are both accurate and feasible under real-world conditions. This study addresses this challenge by evaluating marigold flower detection using three groups of CNN detectors: canonical models, including YOLOv2, Faster R-CNN, and SSD with their original backbones; modified versions of these detectors using DarkNet-53; and modern architectures, including YOLOv11, YOLOv12, and the RT-DETR. The dataset consisted of 392 images from marigold fields, which were manually labeled and augmented to a total of 940 images. The results showed that YOLOv2 with DarkNet-53 achieved the best performance, with 98.8% mean average precision (mAP) and 97.9% F1-score (F1). SSD and Faster R-CNN also improved, reaching 63.1% and 52.8%, respectively. Modern models obtained strong results: YOLOv11 and YOLOv12 reached 96–97%, and RT-DETR 93.5%. The modification of YOLOv2 allowed this classical detector to compete directly with, and even surpass, recent models. Precision–recall (PR) curves, F1-scores, and complexity analysis confirmed the trade-offs between accuracy and efficiency. These findings demonstrate that while modern detectors are efficient baselines, classical models with updated backbones can still deliver state-of-the-art results for UAV-based crop monitoring. Full article

Influenza remains a major health threat due to its high contagiousness and global spread, affecting not only humans but also agricultural livestock and wild animals through transmission via migratory birds. Despite over 70 years of vaccination, influenza still creates epidemics and pandemics, and the ongoing use of vaccination is an essential but currently insufficient strategy. In this study, we assessed the immunogenicity and efficacy of an AP205 virus-like particle (VLP) carrying the HA head domain of the A/PR8/H1N1 strain, administered intranasally and subcutaneously in mice. For this purpose, the entire head region of A/PR8/H1N1 was genetically integrated into a sterically improved version of AP205, which exhibits capsid monomers fused into a dimer, thereby offering inexpensive and scalable production processes. The vaccine induced strong systemic anti-HA IgG and IgA antibodies via both routes, with no significant difference in the levels of IgG. Both immunisation strategies induced protection against a lethal challenge with H1PR8 mouse-adapted influenza virus. The findings demonstrate the potential of the AP205 VLP platform for HA1-based influenza vaccines and its applicability for controlling influenza in both humans and livestock. Full article

This research addresses the challenges of proactive scheduling and routing in manufacturing systems governed by the Material Requirement Planning (MRP) method. Such systems often face capacity constraints, difficulties in resource balancing, and limited traceability of component requirements. The lack of seamless integration between customer orders and production tasks, combined with the manual and time-consuming nature of schedule adjustments, highlights the need for an automated and optimized scheduling method. We propose a novel optimization-based approach that leverages mixed-integer linear programming (MILP) combined with a proprietary procedure for reducing the size of the modeled problem to generate feasible and/or optimal production schedules. The model incorporates dynamic routing, partial resource utilization, limited additional resources (e.g., tools, workers), technological breaks, and time quantization. Key results include determining order feasibility, identifying unfulfilled order components, minimizing costs, shortening deadlines, and assessing feasibility in the absence of available resources. By automating the generation of data from MRP/ERP systems, constructing an optimization model, and exporting the results back to the MRP/ERP structure, this method improves decision-making and competes with expensive Advanced Planning and Scheduling (APS) systems. The proposed innovation solution—the integration of MILP-based optimization with the proprietary PT (data transformation) and PR (model-size reduction) procedures—not only increases operational efficiency but also enables demand source tracking and offers a scalable and economical alternative for modern production environments. Experimental results demonstrate significant reductions in production costs (up to 25%) and lead times (more than 50%). Full article

Patients with ulcerative colitis exhibit an increased risk for supraventricular arrhythmia during the active disease phase of the disease and show signs of atrial electrophysiological remodeling in remission. The goal of this study was to determine the basis for colitis-induced changes in atrial excitability. In a mouse model (C57BL/6; 3 months) of dextran sulfate sodium (DSS)-induced active colitis (3.5% weight/volume, 7 days), electrocardiograms (ECG) revealed altered atrial electrophysiological properties with a prolonged P-wave duration and PR interval. ECG changes coincided with a decreased atrial conduction velocity in Langendorff perfused hearts. Action potentials (AP) recorded from isolated atrial myocytes displayed an attenuated maximal upstroke velocity and amplitude during active colitis, as well as a prolonged AP duration (APD). Voltage clamp analysis revealed a colitis-induced shift in the voltage-dependent activation of the Na-current (I_Na) to more depolarizing voltages. In addition, protein levels of Na_v1.5 protein and connexin isoform Cx43 were reduced. APD prolongation depended on a reduction in the transient outward K-current (I_to) mostly generated by K_v4.2 channels. The changes in ECG, atrial conductance, and APD were reversible upon remission. The change in conduction velocity predominantly depended on the reversibility of the reduced Cx43 and Na_v1.5 expression. Treatment of mice with inhibitors of Angiotensin-converting enzyme (ACE) or Angiotensin II (AngII) receptor type 1 (AT1R) prevented the colitis-induced atrial electrophysiological remodeling. Our data support a colitis-induced increase in AngII signaling that promotes atrial electrophysiological remodeling and puts colitis patients at an increased risk for atrial arrhythmia. Full article

Object detectors often rely on multiple metrics to reflect their accuracy and speed performances independently. This article introduces object detector efficiency index (ODEI), a hardware-agnostic metric designed to assess object detector efficiency based on speed-normalized accuracy, utilizing established concepts including mean average precision (mAP) and floating-point operations (FLOPs). By defining seven mandatory parameters that must be specified when ODEI is invoked, the article aims to clarify long-standing confusions within literature regarding evaluation metrics and promote fair and transparent benchmarking research in the object detection space. Usage demonstration of ODEI using state-of-the-art (SOTA) YOLOv12 and RT-DETRv3 studies is also included. Full article

The root system is vital for Brassica napus water/nutrient uptake and anchorage, highlighting the importance of identifying root development genes (RDGs). In this study, we identified 218 RDGs in B. napus through homology-based retrieval. Phylogenetic analysis of 22 representative species revealed that the RDGs are widely present in plants ranging from aquatic algae to angiosperms. RDGs in B. napus expanded through whole-genome duplication (WGD) events between Brassica rapa and Brassica oleracea ancestors and smaller duplications specific to B. napus. Promoter analysis identified 115 cis-elements, mainly abiotic stress-related and light-responsive. Transcription factor networks showed regulation by BBR-BPC, MIKC_MADS, AP2, and GRAS families. Transcriptome analysis under multiple stresses revealed that low nitrogen (LN) induced the most pronounced changes, with >50% (109/218) of RDGs differentially expressed in roots. Furthermore, we screened the BnaSHR-6 gene, which is co-localized in both primary roots (PR) and lateral roots (LR), and responds strongly to LN. Phenotypic analysis revealed that the BnaSHR-6 gene regulates the growth and development of both PR and LR under LN conditions, and confers a degree of resistance. These findings advance our understanding of RDGs in B. napus and provide valuable gene resources for subsequent molecular breeding. Full article

This study explores the regulatory role of proline-rich acidic protein 1 (Prap1) effects on embryo implantation. A high estrogen model and ovariectomized mice were employed to demonstrate that estrogen regulates Prap1 expression. Uterine tissues were collected from E 1.5 (the presence of a vaginal plug was recorded as embryonal day 0.5, E 0.5) to E 7.5 to detect the Prap1 expression pattern in early pregnancy using qRT-PCR. Embryo adhesion was assessed through uterine perfusion of PRAP1 protein and Prap1 overexpression in endometrial epithelial cells (EECs). The data showed that Prap1 expression was increased in the uterus with high estrogen levels. Prap1 expression was specifically reduced during early implantation. Overexpression of Prap1 in EECs also reduced the embryo adhesions. The differentially expressed genes obtained by RNA-seq were enriched in extracellular matrix and cell adhesion. Muc1 expression was increased in EECs overexpressing Prap1 by RNA-seq and qRT-PCR. Similarly, O-glycosylation biosynthesis was enriched, and glycosylation-related genes were upregulated. Our results demonstrate that Prap1 was regulated by estrogen and an increase in PRAP1 before implantation affected embryo adhesion by regulating the expression of Muc1 and extracellular matrix-related genes, leading to embryo implantation failure. Our results provide a new insight into estrogen regulation of embryo implantation. Full article

Background: Intracranial atherosclerotic stenosis (ICAS) is a leading cause of ischemic stroke, particularly in the anterior circulation. Understanding the underlying stroke mechanisms is essential for guiding personalized treatment strategies. This study proposes an integrated framework that combines CT perfusion imaging, vascular anatomical features, computational fluid dynamics (CFD), and machine learning to classify stroke mechanisms based on the Chinese Ischemic Stroke Subclassification (CISS) system. Methods: A retrospective analysis was conducted on 118 patients with intracranial atherosclerotic stenosis. Key indicators were selected using one-way ANOVA with nested cross-validation and visualized through correlation heatmaps. Optimal thresholds were identified using decision trees. The classification performance of six machine learning models was evaluated using ROC and PR curves. Results: Time to Maximum (Tmax) > 4.0 s, wall shear stress ratio (WSSR), pressure ratio, and percent area stenosis were identified as the most predictive indicators. Thresholds such as Tmax > 4.0 s = 134.0 mL and WSSR = 86.51 effectively distinguished stroke subtypes. The Logistic Regression model demonstrated the best performance (AUC = 0.91, AP = 0.85), followed by Naive Bayes models. Conclusions: This multimodal approach effectively differentiates stroke mechanisms in anterior circulation ICAS and holds promise for supporting more precise diagnosis and personalized treatment in clinical practice. Full article

RAC3 encodes a small GTPase of the Rho family that plays a critical role in actin cytoskeleton remodeling and intracellular signaling regulation. Pathogenic variants in RAC3, all of which reported thus far affect conserved residues within its functional domains, have been linked to neurodevelopmental disorders characterized by diverse phenotypic features, including structural brain anomalies and facial dysmorphism (NEDBAF). Recently, a novel de novo RAC3 variant (NM_005052.3): c.196C>T, p.R66W was identified in a prenatal case with fetal akinesia deformation sequence (a spectrum of conditions that interfere with the fetus’s ability to move), and complex brain malformations featuring corpus callosum agenesis, diencephalosynapsis, kinked brainstem, and vermian hypoplasia. To investigate the mechanisms underlying the association between RAC3 deficiency and this unique, distinct clinical phenotype, we explored the pathophysiological significance of the p.R66W variant in brain development. Biochemical assays revealed a modest enhancement in intrinsic GDP/GTP exchange activity and an inhibitory effect on GTP hydrolysis. Transient expression studies in COS7 cells demonstrated that RAC3-R66W interacts with the downstream effectors PAK1, MLK2, and N-WASP but fails to activate SRF-, AP1-, and NFkB-mediated transcription. Additionally, overexpression of RAC3-R66W significantly impaired differentiation in primary cultured hippocampal neurons. Acute expression of RAC3-R66W in vivo by in utero electroporation resulted in impairments in cortical neuron migration and axonal elongation during corticogenesis. Collectively, these findings suggest that the p.R66W variant may function as an activated version in specific signaling pathways, leading to a distinctive and severe prenatal phenotype through variant-specific mechanisms. Full article

Ventricular pre-excitation (VP) is a cardiac disorder characterized by the presence of an accessory pathway (AP) that bypasses the atrioventricular node (AVN), which, although often asymptomatic, exposes individuals to an increased risk of re-entrant supraventricular tachycardias and sudden cardiac death (SCD) due to rapid atrial fibrillation (AF) conduction. This condition is particularly significant in sports cardiology, where preparticipation ECG screening is routinely performed on athletes. Professional athletes, given their elevated risk of developing malignant arrhythmias, require careful assessment. Early identification of VP and proper risk stratification are crucial for determining the most appropriate management strategy and ensuring the safety of these individuals during competitive sports. Non-invasive tools, such as resting electrocardiograms (ECGs), ambulatory ECG monitoring, and exercise stress tests, are commonly employed, although their interpretation can sometimes be challenging. This review aims to provide practical tips and electrocardiographic clues for detecting VP beyond the classical triad (short PR interval, delta wave, and prolonged QRS interval) and offers guidance on non-invasive risk stratification. Although the diagnostic gold standard remains invasive electrophysiological study, appropriate interpretation of the ECG can help limit unnecessary referrals for young, often asymptomatic, athletes. Full article

Automated segmentation of diabetic retinopathy (DR) lesions is crucial for assessing DR severity and diagnosis. Most previous segmentation methods overlook the detrimental impact of texture information bias, resulting in suboptimal segmentation results. Additionally, the role of lesion shape is not thoroughly considered. In this paper, we propose a lightweight frequency recalibration network (LFRC-Net) for simultaneous multi-lesion DR segmentation, which integrates a frequency recalibration module into the bottleneck layers of the encoder to analyze texture information and shape features together. The module utilizes a Gaussian pyramid to generate features at different scales, constructs a Laplacian pyramid using a difference of Gaussian filter, and then analyzes object features in different frequency domains with the Laplacian pyramid. The high-frequency component handles texture information, while the low-frequency area focuses on learning the shape features of DR lesions. By adaptively recalibrating these frequency representations, our method can differentiate the objects of interest. In the decoder, we introduce a residual attention module (RAM) to enhance lesion feature extraction and efficiently suppress irrelevant information. We evaluate the proposed model’s segmentation performance on two public datasets, IDRiD and DDR, and a private dataset, an ultra-wide-field fundus images dataset. Extensive comparative experiments and ablation studies are conducted across multiple datasets. With minimal model parameters, our approach achieves an mAP_PR of 60.51%, 34.83%, and 14.35% for the segmentation of EX, HE, and MA on the DDR dataset and also obtains excellent results for EX and SE on the IDRiD dataset, which validates the effectiveness of our network. Full article

Estimating the calorie content of fruits is critical for weight management and maintaining overall health as well as aiding individuals in making informed dietary choices. Accurate knowledge of fruit calorie content assists in crafting personalized nutrition plans and preventing obesity and associated health issues. In this paper, we investigate the application of deep learning models for estimating the calorie content in fruits from digital images, aiming to provide a more efficient and accurate method for nutritional analysis. We create a dataset comprising images of various fruits and employ random data augmentation techniques during training to enhance model robustness. We utilize the RT-DETR model integrated into the ultralytics framework for implementation and conduct comparative experiments with YOLOv10 on the dataset. Our results show that the RT-DETR model achieved a precision rate of 99.01% and mAP50-95 of 94.45% in fruit detection from digital images, outperforming YOLOv10 in terms of F1- Confidence Curves, P-R curves, precision, and mAP. Conclusively, in this paper, we utilize a transformer architecture to detect fruits and estimate their calorie and nutritional content. The results of the experiments provide a technical reference for more accurately monitoring an individual’s dietary intake by estimating the calorie content of fruits. Full article

Background: Pemphigus is a group of potentially life-threatening autoimmune bullous diseases induced by pathogenic autoantibodies binding to the surface of epidermal cells. The role of the gut microbiota (GM) has been described in various autoimmune diseases. However, the impact of the GM on pemphigus is less understood. This study aimed to investigate whether there was alterations in the composition and function of the GM in pemphigus patients compared to healthy controls (HCs). Methods: Fecal samples were collected from 20 patients with active pemphigus (AP), 11 patients with remission pemphigus (PR), and 47 HCs. To sequence the fecal samples, 16S rRNA was applied, and bioinformatic analyses were performed. Results: We found differences in the abundance of certain bacterial taxa among the three groups. At the family level, the abundance of Prevotellaceae and Coriobacteriaceae positively correlated with pathogenic autoantibodies. At the genus level, the abundance of Klebsiella, Akkermansia, Bifidobacterium, Collinsella, Gemmiger, and Prevotella positively correlated with pathogenic autoantibodies. Meanwhile, the abundance of Veillonella and Clostridium_XlVa negatively correlated with pathogenic autoantibodies. A BugBase analysis revealed that the sum of potentially pathogenic bacteria was elevated in the AP group in comparison to the PR group. Additionally, the proportion of Gram-negative bacteria in the PR group was statistically significantly lower in comparison to the HC group. Conclusion: The differences in GM composition among the three groups, and the correlation between certain bacterial taxa and pathogenic autoantibodies of pemphigus, support a linkage between the GM and pemphigus. Full article

Background: Acute coronary syndrome (ACS) is a severe cardiovascular disease with globally rising incidence and mortality rates. Traditional risk assessment tools are widely used but are limited due to the complexity of the data. Methods: This study introduces a gated Transformer model utilizing machine learning to analyze electronic health records (EHRs) for an enhanced prediction of major adverse cardiovascular events (MACEs) in ACS patients. The model’s efficacy was evaluated using metrics such as area under the curve (AUC), precision–recall (PR), and F1-scores. Additionally, a patient management platform was developed to facilitate personalized treatment strategies. Results: Incorporating a gating mechanism substantially improved the Transformer model’s performance, especially in identifying true-positive cases. The TabTransformer+Gate model demonstrated an AUC of 0.836, a 14% increase in average precision (AP), and a 6.2% enhancement in accuracy, significantly outperforming other deep learning approaches. The patient management platform enabled healthcare professionals to effectively assess patient risks and tailor treatments, improving patient outcomes and quality of life. Conclusion: The integration of a gating mechanism within the Transformer model markedly increases the accuracy of MACE risk predictions in ACS patients, optimizes personalized treatment, and presents a novel approach for advancing clinical practice and research. Full article