Search Results (7,854)

Background/Objectives: The aim of our study was to identify the most reliable MRI measurements associated with patellar dislocation. Methods: MRI scans from 86 knees (48 controls and 38 with a history of patellar dislocation) were retrospectively analyzed. The following parameters were measured: lateral trochlear inclination (LTI) and its modified version, sulcus angle (SA), trochlear depth (TD), tibial tubercle–trochlear groove distance (TT–TG), patellar tendon–lateral trochlear ridge distance (PT–LTR), and PT–LTR horizontal, a novel modification. Inter-rater reliability was assessed using intraclass correlation coefficients (ICCs), and diagnostic accuracy was evaluated using ROC analysis. Results: All measurements significantly differed between the groups (p < 0.05). SA and TD were highly discriminative (AUC > 0.8) but demonstrated lower inter-rater agreement. PT-LTR horizontal strongly correlated with PT-LTR and was equally sensitive and specific for patellar dislocation as PT-LTR (81.6% and 87.5%, respectively) when in line or extending more laterally than the lateral trochlear ridge (AUC = 0.896, p < 0.001). LTI demonstrated the highest diagnostic performance with a sensitivity of 89.5% and a specificity of 97.9% for a cut-off ≤12.85° (AUC = 0.981), with excellent inter-rater agreement. LTI modified also performed well (AUC = 0.937), with a sensitivity and specificity of 81.6% and 93.7%, respectively. Conclusions: LTI, PT–LTR, and their modified versions demonstrated the highest reliability and diagnostic performance among the MRI measurements evaluated. Given their reproducibility and ease of application, these parameters may be useful in the imaging assessment of patellar dislocation. Further prospective studies are recommended to confirm their clinical utility in broader populations. Full article

Background: The purpose of this study was to investigate the value of machine learning models for preoperative non-invasive prediction of International Society of Urological Pathology (ISUP) grading in clear cell renal cell carcinoma (ccRCC) based on CT urography (CTU)-related peritumoral area (PAT) radiomics features. Methods: We retrospectively analysed 328 ccRCC patients from our institution, along with an external validation cohort of 175 patients from The Cancer Genome Atlas. A total of 1218 radiomics features were extracted from contrast-enhanced CT images, with LASSO regression used to select the most predictive features. We employed four machine learning models, namely, Logistic Regression (LR), Multilayer Perceptron (MLP), Support Vector Machine (SVM), and Extreme Gradient Boosting (XGBoost), for training and evaluation using Receiver Operating Characteristic (ROC) analysis. The model performance was assessed in training, internal validation, and external validation sets. Results: The XGBoost model demonstrated consistently superior discriminative ability across all datasets, achieving AUCs of 0.95 (95% CI: 0.92–0.98) in the training set, 0.93 (95% CI: 0.89–0.96) in the internal validation set, and 0.92 (95% CI: 0.87–0.95) in the external validation set. The model significantly outperformed LR, MLP, and SVM (p < 0.001) and demonstrated prognostic value (Log-rank p = 0.018). Transcriptomic analysis of model-stratified groups revealed distinct biological signatures, with high-grade predictions showing significant enrichment in metabolic pathways (DPEP3/THRSP) and immune-related processes (lymphocyte-mediated immunity, MHC complex activity). These findings suggest that peritumoral imaging characteristics provide valuable biological insights into tumor aggressiveness. Conclusions: The machine learning models based on PAT radiomics features of CTU demonstrated significant value in the non-invasive preoperative prediction of ISUP grading for ccRCC, and the XGBoost modeling had the best predictive ability. This non-invasive approach may enhance preoperative risk stratification and guide clinical decision-making, reducing reliance on invasive biopsy procedures. Full article

Collaborative filtering (CF) continues to be a fundamental approach in recommendation systems for providing users with personalized suggestions. However, such kind of recommender systems are prone to performance issues when faced with noisy, inconsistent, or deliberately manipulated user ratings. Although Generative Adversarial Networks (GANs) offer promising solutions to capture complex user-item interactions in these CF situations, many existing GAN-based methods assume uniform reliability across all ratings, reducing their effectiveness under uncertain conditions. To overcome this challenge, this paper presents DST-AttentiveGAN to introduce a confidence-aware adversarial framework specifically designed to denoise inconsistent ratings in collaborative filtering scenarios. The proposed approach employs Dempster-Shafer Theory (DST) to compute confidence scores by aggregating diverse behavioral indicators, such as item popularity, user activity, and rating variance. These scores guide both components of the GAN architecture in which the generator incorporates a cross-attention mechanism to highlight trustworthy features, while the discriminator uses DST-based confidence to evaluate the credibility of input ratings. Training is carried out using a stabilized Wasserstein GAN objective that promotes both robustness and convergence efficiency. Experimental results in three benchmark data sets show that DST-AttentiveGAN consistently surpasses conventional GAN-based models, delivering more accurate and reliable recommendations under conditions of uncertainty. Full article

Background and Objectives: The Naples Prognostic Score (NPS), a composite index indicative of nutritional and inflammatory status, has been suggested as an important prognostic marker. Uric acid, an indicator of oxidative stress and endothelial impairment, is also associated with cardiovascular risk. This study sought to examine the synergistic value of NPS and uric acid levels in forecasting long-term major adverse cardiovascular and cerebrovascular events (MACCE) in patients with chronic coronary syndrome (CCS) undergoing percutaneous coronary intervention (PCI), using time-varying hazard ratio and time-dependent Receiver Operating Characteristic (ROC) analyses. Materials and Methods: A retrospective analysis was conducted on 288 patients diagnosed with CCS from January 2020 to November 2023. The NPS was determined utilizing serum albumin, total cholesterol, the neutrophil-to-lymphocyte ratio (NLR), and the lymphocyte-to-monocyte ratio (LMR). Cox regression, time-varying hazard ratio models, and time-dependent ROC curve analyses were performed to assess both temporal risk patterns and predictive performance. The principal endpoint was the incidence of MACCE. Results: Major adverse cardiovascular and cerebrovascular events (MACCE) occurred in 69 individuals, representing 23.4% of the total cohort. Both high NPS and elevated uric acid were independently associated with an increased risk of MACCE. The integration of the NPS with uric acid showed superior discriminative and reclassification capabilities compared to the use of each marker independently (p < 0.05 for all). Time-varying hazard ratio analyses demonstrated that the prognostic impact of the NPS was more pronounced in the early follow-up, while the effect of uric acid became stronger in the late phase. Time-dependent ROC analyses confirmed that the combined use of the NPS and uric acid provided superior predictive accuracy compared with either parameter alone across the follow-up period. Conclusions: NPS and uric acid offer complementary prognostic information in CCS. Their combined assessment improves long-term risk stratification, while time-varying and time-dependent analyses reveal that their predictive effects evolve dynamically throughout follow-up. This integrated evaluation may improve clinical decision-making and risk stratification in routine practice. Full article

Objective: Medical Visual Question Answering (Med-VQA), a key technology that integrates computer vision and natural language processing to assist in clinical diagnosis, possesses significant potential for enhancing diagnostic efficiency and accuracy. However, its development is constrained by two major bottlenecks: weak few-shot generalization capability stemming from the scarcity of high-quality annotated data and the problem of catastrophic forgetting when continually learning new knowledge. Existing research has largely addressed these two challenges in isolation, lacking a unified framework. Methods: To bridge this gap, this paper proposes a novel Evolvable Clinical-Semantic Alignment (ECSA) framework, designed to synergistically solve these two challenges within a single architecture. ECSA is built upon powerful pre-trained vision (BiomedCLIP) and language (Flan-T5) models, with two innovative modules at its core. First, we design a Clinical-Semantic Disambiguation Module (CSDM), which employs a novel debiased hard negative mining strategy for contrastive learning. This enables the precise discrimination of “hard negatives” that are visually similar but clinically distinct, thereby significantly enhancing the model’s representation ability in few-shot and long-tail scenarios. Second, we introduce a Prompt-based Knowledge Consolidation Module (PKC), which acts as a rehearsal-free non-parametric knowledge store. It consolidates historical knowledge by dynamically accumulating and retrieving task-specific “soft prompts,” thus effectively circumventing catastrophic forgetting without relying on past data. Results: Extensive experimental results on four public benchmark datasets, VQA-RAD, SLAKE, PathVQA, and VQA-Med-2019, demonstrate ECSA’s state-of-the-art or highly competitive performance. Specifically, ECSA achieves excellent overall accuracies of 80.15% on VQA-RAD and 85.10% on SLAKE, while also showing strong generalization with 64.57% on PathVQA and 82.23% on VQA-Med-2019. More critically, in continual learning scenarios, the framework achieves a low forgetting rate of just 13.50%, showcasing its significant advantages in knowledge retention. Conclusions: These findings validate the framework’s substantial potential for building robust and evolvable clinical decision support systems. Full article

Accurate thickness recognition plays a vital role in safeguarding the structural reliability of critical assets. Pulse eddy current testing (PECT), as a non-destructive method that is both non-contact and insensitive to surface coatings, provides an efficient pathway for this purpose. Nevertheless, the complex, nonstationary, and nonlinear characteristics of PECT signals make it difficult for conventional models to jointly capture localized high-frequency patterns and long-range temporal dependencies, thereby constraining their prediction performance. To overcome these issues, we introduce a novel deep learning framework, multi-scale residual dilated convolution, and bidirectional long short-term memory with a squeeze-and-excitation mechanism (MRDC-BiLSE) for PECT time series analysis. The architecture integrates a multi-scale residual dilated convolution block. By combining dilated convolutions with residual connections at different scales, this block captures structural patterns across multiple temporal resolutions, leading to more comprehensive and discriminative feature extraction. Furthermore, to better exploit temporal dependencies, the BiLSTM-SE module combines bidirectional modeling with a squeeze-and-excitation mechanism, resulting in more discriminative feature representations. Experiments on experimental PECT datasets confirm that MRDC-BiLSE surpasses existing methods, showing applicability for real-world thickness recognition. Full article

With increasingly scarce spectrum resources, frequency-domain signal overlap interference has become a critical issue, making multi-user modulation classification (MUMC) a significant challenge in wireless communications. Unlike single-user modulation classification (SUMC), MUMC suffers from feature degradation caused by signal aliasing, feature redundancy, and low inter-class discriminability. To address these challenges, this paper proposes a collaborative “separation–recognition” framework. The framework begins by separating overlapping signals via a band partitioning and FastICA module to alleviate feature degradation. For the recognition phase, we design a dual-branch network: one branch extracts prior knowledge features, including amplitude, phase, and frequency, from the I/Q sequence and models their temporal dependencies using a bidirectional LSTM; the other branch learns deep hierarchical representations directly from the raw signal through multi-scale convolutional layers. The features from both branches are then adaptively fused using a gated fusion module. Experimental results show that the proposed method achieves superior performance over several baseline models across various signal conditions, validating the efficacy of the dual-branch architecture and the overall framework. Full article

As a key form of deep foundation in civil engineering, the concrete pouring quality of cast-in-place piles directly determines the integrity and long-term bearing performance of the pile body. Accurate monitoring of the pouring interface is critical to preventing defects such as mud inclusion and pile breakage. To address the limitations of existing monitoring methods for concrete pouring interfaces, this paper proposes a full-pulse acoustic monitoring method for the concrete pouring interface of cast-in-place piles. Firstly, by constructing a hardware system platform consisting of “multi-level in-borehole sound sources + interface acoustic wave sensors + orifice full-pulse receivers + ground processors”, differential capture of signals propagating at different depths is achieved through multi-frequency excitation. Subsequently, a waveform data processing method is proposed to realize denoising, enhancement, and frequency discrimination of different signals, and a target feature recognition model that integrates cross-correlation functions and signal similarity analysis is established. Finally, by leveraging the differential characteristics of measurement signals at different depths, a near-field measurement mode and a far-field measurement mode are developed, thereby establishing a calculation model for the elevation position of the pouring interface under different scenarios. Meanwhile, the feasibility of the proposed method is verified through practical engineering cases. The results indicate that the proposed full pulse acoustic monitoring method can achieve non-destructive, real-time, and high-precision monitoring of the pouring interface, providing an effective technical approach for quality control in pile foundation construction and exhibiting broad application prospects. Full article

Background/Objectives: Sodium (Na) concentration and the sodium-to-potassium (Na/K) ratio in human milk reflect epithelial tight junction integrity and have been proposed as non-invasive biomarkers of lactational dysfunction, including subclinical mastitis and ductal obstruction. However, their discriminative performance across varied mammary duct conditions, as well as their relevance to milk quality and nutritional integrity, remain underexplored. This study aimed to evaluate the ability of Na, K and the Na/K ratio to discriminate ductal obstruction from non-obstructed lactation—including normal, mixed, and donor milk—and to assess their applicability as nutritional and clinical screening biomarkers. Methods: The study analyzed 635 human milk samples from four groups: obstructed ducts (n = 94), mixed ducts (n = 39), normal ducts (n = 102), and donor milk (n = 400). Na and K concentrations were measured using validated handheld ion-selective electrode analyzers. Statistical analyses included Quade’s ANCOVA and receiver operating characteristic curve analysis, adjusting for infant age, gestational age, birth body weight, maternal age and storage duration. Results: Na concentrations were highest in obstructed ducts (Group A: median 810 ppm, IQR 368–1725) compared with normal ducts (Group C: 220 ppm, IQR 140–283) and donor milk (Group D: 98 ppm, IQR 80–130) (p < 0.001). A similar pattern was observed for the Na/K ratio (Group A: 1.5, IQR 0.6–3.1 vs. Group C: 0.3, IQR 0.2–0.5; Group D: 0.3, IQR 0.2–0.3). After adjusting, both Na and the Na/K ratio remained significantly elevated in milk from obstructed ducts compared to non-obstructed samples (p < 0.001). Donor milk exhibited the lowest and most stable electrolyte levels. Na demonstrated excellent discriminative performance (area under the curve = 0.96), slightly outperforming the Na/K ratio (area under the curve = 0.92). Conclusions: Na concentration and the Na/K ratio in human milk are sensitive and practical biomarkers of mammary gland integrity. Given that Na alone can be measured without additional calculations, its simplicity and strong performance support its application as a potential biomarker for ductal obstruction, with implications for both lactation support and nutritional science. Full article

To meet the demand for energy-efficient and high-performance computing in resource-limited sensor edge applications, this paper presents a reconfigurable memristor-based computing-in-memory circuit for Content-Addressable Memory (CAM). The scheme exploits the analog multi-level resistance characteristics of memristors to enable parallel multi-bit processing, overcoming the constraints of traditional binary computing and significantly improving storage density and computational efficiency. Furthermore, by employing dynamic adjustment of the mapping between input signals and reference voltages, the circuit supports dynamic switching between exact and approximate CAM modes, substantially enhancing functional flexibility. Experimental results demonstrate that the 32 × 36 memristor array based on a TiN/TiOx/HfO₂/TiN structure exhibits eight stable and distinguishable resistance states with excellent retention characteristics. In large-scale array simulations, the minimum voltage separation between state-representing waveforms exceeds 6.5 mV, ensuring reliable discrimination by the readout circuit. This work provides an efficient and scalable hardware solution for intelligent edge computing in next-generation sensor networks, particularly suitable for real-time biometric recognition, distributed sensor data fusion, and lightweight artificial intelligence inference, effectively reducing system dependence on cloud communication and overall power consumption. Full article

To address the limitations of single indices in comprehensively evaluating the quality of Korla fragrant pears, this study proposes the firmness–soluble solids ratio (FSR), defined as the ratio of average firmness (FI) to soluble solid content (SSC) for each individual fruit, as a novel index. Using 600 samples from five maturity stages with hyperspectral imaging (950–1650 nm), the dataset was split 4:1 by the SPXY algorithm. The findings demonstrated that FSR’s effectiveness in quantifying the dynamic relationship between FI and SSC during maturation. The developed multiscale convolutional neural network–long short-term memory (MSCNN–LSTM) model achieved high prediction accuracy with determination coefficients of 0.8934 (FI), 0.8731 (SSC), and 0.8610 (FSR), and root mean square errors of 0.9001 N, 0.7976%, and 0.1676, respectively. All residual prediction deviation values exceeded 2.5, confirming model robustness. The MSCNN–LSTM showed superior performance compared to other benchmark models. Furthermore, the integration of prediction models with visualization techniques successfully mapped the spatial distribution of quality indices. For maturity discrimination, hyperspectral-based partial least squares discriminant analysis and linear discriminant analysis models achieved perfect classification accuracy (100%) under five-fold cross-validation across all five maturity stages. This work provides both a theoretical basis and a technical framework for non-destructive evaluation of comprehensive quality and maturity in Korla fragrant pears. Full article

Photon-counting computed tomography (PCCT) represents a major technological innovation compared to conventional CT, offering improved spatial resolution, reduced electronic noise, and intrinsic spectral capabilities. These advances open new perspectives for synergy with radiomics, a field that extracts quantitative features from medical images. The ability of PCCT to generate multiple types of datasets, including high-resolution conventional images, iodine maps, and virtual monoenergetic reconstructions, increases the richness of extractable features and potentially enhances radiomics performance. This narrative review investigates the current evidence on the interplay between PCCT and radiomics in cardiothoracic imaging. Phantom studies demonstrate reduced reproducibility between PCCT and conventional CT systems, while intra-scanner repeatability remains high. Nonetheless, PCCT introduces additional complexity, as reconstruction parameters and acquisition settings significantly may affect feature stability. In chest imaging, early studies suggest that PCCT-derived features may improve nodule characterization, but existing machine learning models, such as those applied to interstitial lung disease, may require recalibration to accommodate the new imaging paradigm. In cardiac imaging, PCCT has shown particular promise: radiomic features extracted from myocardial and epicardial tissues can provide additional diagnostic insights, while spectral reconstructions improve plaque characterization. Proof-of-concept studies already suggest that PCCT radiomics can capture myocardial aging patterns and discriminate high-risk coronary plaques. In conclusion, evidence supports a growing synergy between PCCT and radiomics, with applications already emerging in both lung and cardiac imaging. By enhancing the reproducibility and richness of quantitative features, PCCT may significantly broaden the clinical potential of radiomics in computed tomography. Full article

Background/Objectives: Estimating the sex from unknown individuals is a critical step when constructing their biological profile. The distal humerus is a useful sex discriminator as shown through metric, morphoscopic, and geometric morphometric approaches. A recently developed web application using geometric morphometric techniques has provided an accessible tool for estimating sex from the shape of the olecranon fossa. The aims of this study were to examine the accuracy of the Ammer–Coelho web application on Scottish individuals, as well as test its repeatability and reproducibility among seven different observers. Methods: The right humerus was obtained from 52 Scottish individuals, and the Ammer–Coelho web application was used to estimate sex. Total accuracy rates and sex-specific rates were calculated, and an analysis of Cohen’s and Fleiss’ kappa was performed. Results: The results demonstrate an overall accuracy of 69.23% with a sex bias of −5.33%, with 55.56% of the sample being accurately estimated with probabilities equal to or higher than 0.95. Substantial agreement was reported for intra-observer error, and an overall low agreement was reported for inter-observer error Conclusions: This is the first study that evaluates the Ammer–Coelho web application. A tendency to perceive more triangular shapes (male appearance) rather than oval shapes (female appearance) resulted in a high level of observer errors, with only 6% of females correctly estimated across the seven observers. The low accuracy rates obtained could also indicate inter-population variation, as shown by other studies. Due to the results obtained, research considering different levels of observers’ experience and diverse population samples is needed to confirm our findings. Full article

Hypoxis hemerocallidea (Hypoxidaece) is thoroughly researched and well documented for its plethora of anecdotal and scientifically backed pharmacological potentials. Its anecdotal uses and pharmacological activities are attributed to its extract’s inherent bioactive compounds like hypoxoside, rooperol, and β-sitosterol. This study aimed at conducting a targeted and holistic phytochemical profiling of variations in Hypoxis hemerocallidea (H. hemerocallidea) and related species. The chemotaxonomic classifications of H. hemerocallidea and seven other related species were also carried out to avert the possibility of over harvesting H. hemerocallidea and the encouragement of species inter-change. The plant extracts were analysed with reverse phase ultra-pure liquid chromatography quadrupole time-of-flight mass spectrometry and gas chromatography, as well as high-performance thin-layer chromatography. The generated chromatographic data were made compatible for chemometric computation using Principal Component Analysis (PCA) and Orthogonal Partial Least Squares Discriminant Analysis (OPLS-DA) models. The results obtained unveil orcinol glycoside, curculigoside C, hypoxoside, dehydroxyhypoxoside, bisdehydroxy hypoxoside, hemerocalloside, galpinoside, cholchicoside, geraniol glycoside, β-sitosterol, oleic acid, and 2-hydroxyethyl linoleate as target phytochemicals that define the profiles of the Hypoxis species. In addition, three distinct chemotypes defined by hemerocalloside, galpinoside, and colchicoside, respectively, were observed, as well as holistic variations in all secondary metabolites. Due to similarities in the phytochemical constituents of selected species, species inter-change seems imminent if further research confirms the findings of this study. Full article

Background: Hypertrophic cardiomyopathy (HCM) is associated with sudden cardiac death (SCD). The HCM Risk-SCD model has been widely used in Western populations, but its performance in Chinese patients remains unclear. Methods: This retrospective cohort study evaluated 534 HCM patients (348 males and 186 females) at Fuwai Hospital from 1992 to 2010. We calculated the HCM Risk-SCD score for each patient and categorized them into low-risk (<4%) and intermediate–high-risk (≥4%) groups. The primary endpoint was SCD events, defined as unexpected sudden death within one hour of symptom onset, successful resuscitation after cardiac arrest, appropriate ICD discharge, or sustained ventricular tachycardia. Model performance was assessed using Cox regression analysis, Kaplan–Meier survival analysis, ROC curve analysis, and subgroup analyses with interaction tests. Results: During a mean follow-up of 6.96 ± 4.16 years, 31 SCD events occurred. The intermediate–high-risk group had significantly higher SCD incidence than the low-risk group (8.68% vs. 3.42%, p = 0.01). This association remained significant after multivariate adjustment (HR 2.718, 95% CI: 1.264–5.848, p = 0.011). Kaplan–Meier analysis showed significant differences in SCD-free survival between risk strata (log-rank p = 0.01). The actual 5-year SCD event rate (4.31%) closely aligned with the model-predicted rate (4.65 ± 3.26%). ROC analysis demonstrated moderate discriminative ability in the overall population (AUC = 0.660, p = 0.003). The optimal cutoff value was 3.23 for the overall population. Conclusions: The HCM Risk-SCD model demonstrates acceptable performance in Chinese HCM patients. Full article