Search Results (1,965)

Modern external beam radiotherapy (EBRT) on lung cancer improved dose distribution thanks to advanced dose calculation algorithms, but side effects and relapses can occur in any case onset. Differential diagnosis of relapses and side effects is difficult, and when computed tomography (CT) is uncertain 18-fluorodeoxyglucose-positron emission tomography/computed tomography (18F-FDG-PET/CT) can support the diagnosis, even if it can also be difficult to construe. The aim of this retrospective analysis was to evaluate 18F-FDG PET/CT qualitative patterns and semiquantitative parameters, both automatic and preceded by physicians, in interpreting lung lesions in the radiotherapy (RT) lung irradiation field. In total, 94 patients (pts) submitted to EBRT (3 months before) for stage II lung cancer were included (74 men, 20 women, mean age of 68 years old, range of 49–84 years old). CT scans were performed on pts, which showed lung lesions in the RT field. 18F-FDG-PET/CT scans were analyzed qualitatively as negative or positive, and the presence of the lung area with a high 18F-FDG uptake pattern was distinguished as the following: focal/wide, deep/shade, or homogeneous/inhomogeneous. Furthermore, the following semiquantitative parameters were collected: gSUVmax (global standardized uptake value max), MTV (tumor metabolic volume), metabolic spatial distribution (MSD) = proximal SUVmax/distal SUVmax, and intratumoral difference in spatial distribution (IDSD%) = [distal SUVmax/proximal SUVmax] × 100. 18F-FDG PET/CT was related to the pts’ outcome (biopsy and/or clinical–instrumental follow-up): positive for lung relapse, negative if the lesions were phlogistic. The following diagnostic performance parameters of 18F-FDG PET/CT were calculated: sensitivity (Sens), specificity (Spec), diagnostic accuracy (DA), positive predictive value (PPV), and negative predictive value (NPV). Qualitative variables were compared by Chi-squared test, while for semiquantitative parameters Student’s t-test was applied; p < 0.05 was considered statistically significant. Statistics tests were performed with MedCalc V.22.018 ©2024. In 76/94 (80.8%) pts, 18F-FDG uptake was higher compared to the background; in 18/94 (19.2%) no high 18F-FDG uptake areas were detected. Outcome was positive for lung relapse in 49/94 pts, while negative in 45/94, with disease prevalence of 52.13% (95%CI = 41.57–62.54%). In the 18/94 pts without high 18F-FDG uptake, the outcome was negative for lung relapse. In 49/76 pts with higher 18F-FDG uptake, the outcome confirmed the presence of relapse, while in 27/76 the lesion was phlogistic. Results about the Sens, Spec, DA, PPV, and NPV (95%CI) were, respectively: 100% (92.75–100%), 40% (25.7–55.67%), 71.28% (61.02–80.14%), 64.47% (58.84–69.73%), and 100% (81.47–100%). Chi-square test showed significant statistical difference between the positive and negative outcome for patterns focal/wide (p = 0.02) and deep/shade (p < 0.00001). A total of 35/49 (71.4%) pts with lung relapse had a focal lesion and 15/27 (55.6%) with phlogosis had a wide pattern. A total of 34/49 (69.4%) pts with lung relapse had a deep pattern and 25/27 (92.6%) with lung phlogosis had the shade one. Significant difference was observed in evaluating the three patterns (p = 0.00007), with prevalence of “focal/deep/homogeneous” patterns in lung relapse and “wide/shade/inhomogeneous” in phlogosis. gSUVmax, MTV, MSD, and IDSD% were in the following order: in the 76 pts, 5.63 (1.4–24.7), 42.49 (4.94–193), 3.61 (1–5.54), and 70.7% (18–100%); in the 49/76 true positive pts, 6.93 (1.5–24.7), 35.28 (4.94–85.99), 3.30 (1.05–5.54), and (18–95%); in the 27/76 false positive pts, 3.27 (1.4–19.2), 38.37 (4.94–193), 1.57 (1–2.13), and 78.6% (4.7–100%). The difference was statistically significant only for MSD (t = 2.779; p = 0.0069) and IDSD% (t = 2.769; p = 0.0071). 18F-FDG-PET/CT confirms its high sensitivity and NPV in evaluating lung lesions after RT. To improve physician confidence in interpreting lung 18F-FDG uptake without further support, MSD and IDSD% could be considered. Heterogeneity of lung lesions, especially in radiotreated tissue, can be turned from a drawback to a resource and analyzed for differentiating relapses from EBRT side effects. Considering the calculation of semiquantitative parameters that require “human intelligence”, even if slightly more time-consuming, can improve the nuclear physician’s confidence in interpreting 18F-FDG PET/CT images. Full article

Loss functions play a significant role in shaping model behavior in machine learning, yet their design implications remain underexplored in natural language processing tasks such as Named Entity Recognition (NER). This study investigates the performance and optimization behavior of five loss functions—L1, L2, Cross-Entropy (CE), KL Divergence (KL), and the proposed DLITE (Discounted Least Information Theory of Entropy) Loss—within transformer-based NER models. DLITE introduces a bounded, entropy-discounting approach to penalization, prioritizing recall and training stability, especially under noisy or imbalanced data conditions. We conducted empirical evaluations across three benchmark NER datasets: Basic NER, CoNLL-2003, and the Broad Twitter Corpus. While CE and KL achieved the highest weighted F1-scores in clean datasets, DLITE Loss demonstrated distinct advantages in macro recall, precision–recall balance, and convergence stability—particularly in noisy environments. Our findings suggest that the choice of loss function should align with application-specific priorities, such as minimizing false negatives or managing uncertainty. DLITE adds a new dimension to model design by enabling more measured predictions, making it a valuable alternative in high-stakes or real-world NLP deployments. Full article

Background/Objectives: The diagnosis of prosthetic joint infection remains difficult. Microbiological cultures frequently have false-positive and false-negative results. This study investigates whether rapid nanopore sequencing can be used to aid the identification of bacteria causing prosthetic joint infection for more timely identification and treatment. Methods: Nineteen patients who had revision surgery following total joint arthroplasty were included in this study. Of these, 15 patients had an infected joint arthroplasty. All patients had joint fluid aspirated at the time of revision surgery. The DNA was extracted from these fluid aspirates, and rapid nanopore sequencing was performed using the MinION device from Oxford Nanopore Technologies. The sequencing data was trimmed to improve quality and filtered to remove human reads using bioinformatic tools. Genomic sequence classification was performed using the Basic Local Alignment Search Tool. The results were filtered by read length and sequence identity score. The European Bone and Joint Infection Society criteria were used as a standard to identify infected and not infected patients. Confusion tables were used to calculate accuracy and F1 score based on this criteria and the nanopore sequencing results. Results: Microbiological cultures and nanopore sequencing had an accuracy of 68% and 74%, respectively. However, combining both results predicted infection accurately in 94% of cases (F1 score 96%). Conclusions: Nanopore sequencing has the potential to aid identification of bacteria causing prosthetic joint infection and may be useful as a supplementary diagnostic tool. Full article

The aim of this study was to improve screening efficiency by establishing reasonable interpretation criteria for the use of flow injection analysis tandem mass spectrometry (FIA-MS/MS) in newborn screening (NBS) for X-linked adrenoleukodystrophy (X-ALD). FIA-MS/MS was employed to analyze very-long-chain acylcarnitines (ACs) and lysophosphatidylcholines (LPCs) and their ratios in dried blood spot (DBS) obtained from five X-ALD patients in the neonatal period (0–7 days old) and 7123 healthy neonate controls. By comparing these results and analyzing receiver operating characteristic (ROC) curves, we identified sensitive indicators for X-ALD screening in newborns. To evaluate the performance of different FIA-MS/MS screening indicators, we simultaneously analyzed 7712 neonatal DBS samples obtained for X-ALD screening using FIA-MS/MS and the established liquid chromatography tandem mass spectrometry (LC-MS/MS) method for quantitative detection of C26:0-lysophosphatidylcholine (C26:0-LPC). Furthermore, 84,268 newborn X-ALD screening results were retrospectively analyzed to further evaluate the screening performance of FIA-MS/MS. After the three-step optimization evaluation, the optimized first-tier sensitive screening indicators of FIA-MS/MS were C24:0-AC, C26:0LPC, and C24:0/C22:0-AC. Among the 7712 newborns screened, one case was confirmed to be double-positive. Within separate statistical analyses, based on LC-MS/MS screening alone (positive cutoff > 0.17 µmol/L), only seven cases (0.09%) were initially positive, with a positive predictive value (PPV) of 42.8%, and two additional ABCD1 VUS hemizygous males were detected. Through the retrospective analysis of 84,268 newborns, eight ABCD1 variants (six hemizygous males and two heterozygous females) were ultimately identified. Our study showed that the optimization of first-tier screening performance is particularly important if second-tier screening is not performed. Using LC-MS/MS for second-tier screening for X-ALD can significantly reduce the number of false positives, but the method still misses some false negatives. If it is used as a first-tier assessment, more VUS variant neonates can be detected. Full article

The problem of geolocating Reddit users without access to the author information API is tackled in this study. Using subreddit data, we analyzed and identified user location based on their interactions within location-specific subreddits. Using unsupervised learning methods such as Latent Dirichlet Allocation (LDA) and Non-Negative Matrix Factorization (NMF) algorithms, we examined conversations about COVID-19 and immunization across the U.S., focusing on COVID-19 vaccination. Our topic modeling identifies four themes: humor and sarcasm (e.g., jokes about microchips), conspiracy theories (e.g., tracking devices and microchips in the COVID-19 vaccine), public skepticism (e.g., debates over vaccine safety and freedom), and vaccine brand concerns (e.g., Pfizer, Moderna, and booster shots). Our geolocation analysis shows that regions with lower vaccination rates often exhibit a higher prevalence of misinformation-labeled comments. For example, counties such as Ada County (Idaho), Newton County (Missouri), and Flathead County (Montana) showed both a low vaccine uptake and a high rate of false information. This study provides useful information on the many different examples of misinformation that are disseminated online. It gives us a better understanding of how people in different parts of the U.S. think about getting a COVID-19 vaccine. Full article

Identifying homologous vulnerabilities across diverse IoT firmware images is critical for large-scale vulnerability auditing and risk assessment. However, existing approaches often rely on coarse-grained components or single-dimensional metrics, lacking the semantic granularity needed to capture cross-firmware vulnerability relationships. To address this gap, we propose FirmVulLinker, a semantic profiling framework that holistically models firmware images across five dimensions: unpacking signature sequences, filesystem semantics, interface exposure, boundary binary symbols, and sensitive parameter call chains. These multi-dimensional profiles enable interpretable similarity analysis without requiring prior vulnerability labels. We construct an evaluation dataset comprising 54 Known Defective Firmware (KDF) images with 74 verified vulnerabilities and assess FirmVulLinker across multiple correlation tasks. Compared to state-of-the-art techniques, FirmVulLinker achieves higher precision with substantially lower false-positive and false-negative rates. Notably, it identifies and reproduces 53 previously undisclosed N-day vulnerabilities in firmware images not listed as affected at the time of public disclosure, effectively extending the known impact scope. Our results demonstrate that FirmVulLinker enables scalable, high-fidelity homologous vulnerability analysis, offering a new perspective on understanding cross-firmware vulnerability patterns in the IoT ecosystem. Full article

Compounds, which rely on metabolism to exhibit toxicity, pose a challenge for next-generation risk assessment (NGRA). Since many of the currently available non-animal new approach methods (NAMs) lack metabolic activity, their use may lead to an underestimation of the true hazard to humans (false negative predictions). We explored here strategies to deal with metabolite-mediated toxicity in assays for developmental neurotoxicity. First, we present an overview of substances that may serve as potential positive controls for metabolite-related neurotoxicity. Then, we demonstrate, using the MitoMet (UKN4b) assay, which assesses the adverse effects of chemicals on neurites of human neurons, that some metabolites have a higher toxic potency than their parent compound. Next, we designed a strategy to integrate elements of xenobiotic metabolism into assays used for (developmental) neurotoxicity testing. In the first step of this approach, hepatic post-mitochondrial fractions (S9) were used to generate metabolite mixtures (“metabolisation module”). In the second step, these were applied to a NAM (exemplified by the UKN4b assay) to identify metabolite-mediated toxicity. We demonstrate the applicability and transferability of these approaches to other assays, by an exemplary study on the basis of the cMINC (UKN2) assay, another NAM of the developmental neurotoxicity in vitro battery. Based on the experience gained from these experiments, we discuss key issues to be addressed if this approach is to be used more broadly for NAM in the NGRA context. Full article

Bank transaction fraud detection has emerged as an important area of research in the economic sector, driven by the developing sophistication of fraudulent activities and the considerable economic losses they entail. This paper reviews numerous methodologies and technologies employed in the real-time identification and mitigation of fraudulent transactions, including traditional statistical techniques, machine learning algorithms and advanced artificial intelligence strategies. It enhances the need to combine anomaly detection structures with behavioral analytics to enhance detection accuracy while addressing challenges like data privacy, the need to balance false positives and negatives and the need for adaptive systems. By evaluating the most recent developments and case studies, this study provides a comprehensive assessment of what is happening in bank transaction fraud detection and presents future directions for enhancing safety features. Full article

Complex event detection (CED) adds value to camera stream data in various applications such as workplace safety, task monitoring, security, and health. Recent CED frameworks have addressed the issues of limited spatiotemporal labels and costly training by decomposing the CED into low-level features, as well as spatial and temporal relationship extraction. However, these frameworks suffer from high resource costs, low scalability, and an increased number of false positives and false negatives. This paper proposes GICEDCAM, which distributes CED across edge, stateless, and stateful layers to improve scalability and reduce computation cost. Additionally, we introduce a Spatial Event Corrector component that leverages geospatial data analysis to minimize false negatives and false positives in spatial event detection. We evaluate GICEDCAM on 16 camera streams covering four complex events. Relative to a strong open-source baseline configured for our setting, GICEDCAM reduces end-to-end latency by 36% and total computational cost by 45%, with the advantage widening as objects per frame increase. Among corrector variants, Bayesian Network (BN) yields the lowest latency, Long Short-Term Memory (LSTM) achieves the highest accuracy, and trajectory analysis offers the best accuracy–latency trade-off for this architecture. Full article

An unsupervised anomaly-detection framework is proposed and field validated for thrust-bearing monitoring in the impulse turbine of a shoreline oscillating water-column (OWC) wave energy converter (WEC) off Jeju Island, Korea. Operational monitoring is constrained by nonstationary sea states, scarce fault labels, and low-rate supervisory logging at 20 Hz. To address these conditions, a 24 h period of normal operation was median-filtered to suppress outliers, and six physically motivated time-domain features were computed from triaxial vibration at 10 s intervals: absolute mean; standard deviation (STD); root mean square (RMS); skewness; shape factor (SF); and crest factor (CF, peak divided by RMS). A feature-based autoencoder was trained to reconstruct the feature vectors, and reconstruction error was evaluated with an adaptive threshold derived from the moving mean and moving standard deviation to accommodate baseline drift. Performance was assessed on a 2 h test segment that includes a 40 min simulated fault window created by doubling the triaxial vibration amplitudes prior to preprocessing and feature extraction. The detector achieved accuracy of 0.99, precision of 1.00, recall of 0.98, and F1 score of 0.99, with no false positives and five false negatives. These results indicate dependable detection at low sampling rates with modest computational cost. The chosen feature set provides physical interpretability under the 20 Hz constraint, and denoising stabilizes indicators against marine transients, supporting applicability in operational settings. Limitations associated with simulated faults are acknowledged. Future work will incorporate long-term field observations with verified fault progressions, cross-site validation, and integration with digital-twin-enabled maintenance. Full article

The implementation of newborn screening (NBS) has revolutionized the diagnostic landscape of cystic fibrosis (CF). In Portugal, NBS was initiated in October 2013 through a pilot study and was subsequently fully integrated into a nationwide program by December 2018. Infants with positive screening results are referred to a specialized CF reference center for diagnostic confirmation, employing Sweat Chloride Testing (SCT) and genetic testing for CFTR variants. We aimed to analyze infants with a positive CF screening and determine the false positive and false negative rates, as well as to calculate the positive predictive value and sensitivity of our NBS program. A retrospective nationwide analysis was conducted on infants with a positive NBS for CF between October 2013 and February 2023. Two hundred and forty infants were referred from the NBS program; 74 (30.8%) were confirmed to have CF through SCT and genetic testing. Sensitivity was 93.2%, and the positive predictive value (PPV) was 30.8%. In addition, 48.5% were homozygous for F508del variants, and 87.8% had at least one F508del variant. Guidelines set forth by the European Cystic Fibrosis Society advise NBS programs to achieve a minimum PPV of 30% and a minimum sensitivity of 95%. Our report demonstrated good compliance with these recommendations. Full article

To address the detection of stealthy False Data Injection Attacks (FDIA) that evade traditional detection mechanisms in smart grids, this paper proposes an unsupervised learning framework named SHAP-LUNAR (SHapley Additive ExPlanations-Learnable Unified Neighborhood-based Anomaly Ranking). This framework overcomes the limitations of existing methods, including parameter sensitivity, inefficiency in high-dimensional spaces, dependency on labeled data, and poor interpretability. Key contributions include (1) constructing a lightweight k-nearest neighbor graph through learnable graph aggregation to unify local anomaly detection, significantly reducing sensitivity to core parameters; (2) generating negative samples via boundary uniform sampling to eliminate dependency on real attack labels; (3) integrating SHAP for quantifying feature contributions to achieve feature-level model interpretation. Experimental results on IEEE 14-bus and IEEE 118-bus systems demonstrate F1 scores of 99.40% and 96.79%, respectively, outperforming state-of-the-art baselines. The method combines high precision, strong robustness, and interpretability. Full article

Pedestrian accidents involving heavy vehicles such as trucks and buses remain a critical safety issue, primarily due to structural blind spots. While existing systems like radar-based FCW and BSD have been adopted, they are not fully optimized for pedestrian detection, particularly under adverse weather conditions. This study focused on the empirical validation of a 360-degree pedestrian collision avoidance system using multichannel ultrasonic sensors specifically designed for heavy-duty vehicles. Eight sensors were strategically positioned to ensure full spatial coverage, and scenario-based field experiments were conducted under controlled rain (50 mm/h) and fog (visibility <30 m) conditions. Pedestrian detection performance was evaluated across six distance intervals (50–300 cm) using indicators such as mean absolute error (MAE), coefficient of variation (CV), and false-negative rate (FNR). The results demonstrated that the system maintained average accuracy of 97.5% even under adverse weather. Although rain affected near-range detection (FNR up to 17.5% at 100 cm), performance remained robust at mid-to-long ranges. Fog conditions led to lower variance and fewer detection failures. These empirical findings demonstrate the system’s effectiveness and robustness in real-world conditions and emphasize the importance of evaluating both distance accuracy and detection reliability in pedestrian safety applications. Full article

Non-cooperative spatial target detection plays a vital role in enabling autonomous on-orbit servicing and maintaining space situational awareness (SSA). However, due to the limited computational resources of onboard embedded systems and the complexity of spaceborne imaging environments, where spacecraft images often contain small targets that are easily obscured by background noise and characterized by low local information entropy, many existing object detection frameworks struggle to achieve high accuracy with low computational cost. To address this challenge, we propose YOLO-GRBI, an enhanced detection network designed to balance accuracy and efficiency. A reparameterized ELAN backbone is adopted to improve feature reuse and facilitate gradient propagation. The BiFormer and C2f-iAFF modules are introduced to enhance attention to salient targets, reducing false positives and false negatives. GSConv and VoV-GSCSP modules are integrated into the neck to reduce convolution operations and computational redundancy while preserving information entropy. YOLO-GRBI employs the focal loss for classification and confidence prediction to address class imbalance. Experiments on a self-constructed spacecraft dataset show that YOLO-GRBI outperforms the baseline YOLOv8n, achieving a 4.9% increase in mAP@0.5 and a 6.0% boost in mAP@0.5:0.95, while further reducing model complexity and inference latency. Full article

Given the critical importance of timely forest fire detection to mitigate environmental and socio-economic consequences, this research aims to achieve high detection accuracy while maintaining real-time operational efficiency, with a particular focus on minimizing computational complexity. We propose a novel framework that systematically integrates normalization, feature selection, adaptive oversampling, and classifier optimization to enhance detection performance while minimizing computational overhead. The evaluation is conducted using three distinct Canadian forest fire datasets: Alberta Forest Fire (AFF), British Columbia Forest Fire (BCFF), and Saskatchewan Forest Fire (SFF). Initial classifier benchmarking identified the best-performing tree-based model, followed by normalization and feature selection optimization. Next, four oversampling methods were evaluated to address class imbalance. An ablation study quantified the contribution of each module to overall performance. Our targeted, stepwise strategy eliminated the need for exhaustive model searches, reducing computational cost by 97.75% without compromising accuracy. Experimental results demonstrate substantial improvements in F1-score, AFF (from 69.12% to 82.75%), BCFF (61.95% to 77.91%), and SFF (90.03% to 96.18%) alongside notable reductions in False Negative Rates compared to baseline models. Full article