Search Results (161)

Newborn screening (NBS) for spinal muscular atrophy (SMA) and severe combined immunodeficiency (SCID) faces challenges. Accurate and precise SMN1 and SMN2 copy number determination, confirmed by two orthogonal methods, are vital for SMA prognostication and treatment. Single SMN1 copy detection also enables the further feasibility to screen for compound heterozygotes. In SCID, low-level T-cell receptor excision circle (TREC) quantification by quantitative PCR is imprecise, necessitating replicates for reliable results. An assay with enhanced accuracy, precision, and high throughput is warranted for NBS SMA and SCID. False positive of SMN1 deletions due to allele dropout are also a potential pitfall in PCR-based methods. We evaluated a first-tier fully automated quadruplex droplet digital PCR (ddPCR) assay detecting SMN1, SMN2, TREC, and RPP30 using dried blood spots together with a second-tier Sanger sequencing to exclude SMN1 allele dropout. Five proficiency test samples and six patient samples with known SMN1 and SMN2 copy numbers confirmed by multiplex ligation-dependent probe amplification were used for accuracy evaluation with full concordance. The ddPCR assay showed high precision for SMN1 and SMN2 (<7% coefficient of variation (CV) for ≥0 copy) and TREC (14.6% CV at 37 copies/µL blood). Second-tier Sanger sequencing identified all SMA cases with homozygous deletions. Accuracy for TREC classification was concordant with 10 proficiency samples. The reference interval of TREC concentration was established for newborns ≥ 34 weeks (n = 1812) and the 2.5th percentile was 57 copies/µL blood. A two-tiered approach with fully automated quadruplex ddPCR and Sanger sequencing delivers accurate and precise quantitation for NBS SMA and SCID, enabling early treatment and counseling. Full article

Introduction: Lumbar radiculopathies involving the entrapment of nerve roots in the lumbar spine are common neuropathic conditions. These conditions affect 40% to 70% of individuals in their lifetime and lead to significant medical costs. Objective: This study aims to identify clinical, psychological, and biomarker-based prognostic factors that predict functional outcomes following surgery for lumbar radiculopathy. Materials and Methods: This prospective cohort study, conducted at Hospital Central de la Defensa Gómez Ulla, Madrid (Spain), adheres to the STROBE guidelines. The study includes patients aged 18–75 with lumbar radiculopathy, confirmed by clinical diagnosis, imaging, and electromyography (EMG) findings. Exclusion criteria include previous lumbar spine surgeries and systemic diseases. The primary outcome is the Oswestry Low Back Pain Disability Questionnaire. Sample size calculations, based on a conservative effect size (f² = 0.20), determined the need for 172 participants, accounting for a 15% dropout rate and 80% power. Procedure: Patients will undergo an initial assessment, including EMG tests, sociodemographic and psychological questionnaires, blood sample tests, and physical questionnaires. This process will be repeated six months post-intervention, except for the blood sample test, expectations questionnaire, and EMG, which will be performed only once. Statistical Analyses: Data will be analyzed using Python 3.12.3, employing a multivariate linear regression analysis. Assumptions of linearity, independence, homoscedasticity, normality, and no multicollinearity will be validated. Corrective measures will be applied if assumptions are violated. Ethics and Dissemination: The study follows the Declaration of Helsinki guidelines and has been approved by the Ethics Committee of Universidad Rey Juan Carlos (070220241052024). Potential risks will be minimized, and adverse events will be recorded and addressed. Findings will be published in high-impact journals and presented at conferences. Full article

Mine water hazards represent one of the principal threats to safe coal mine operations; therefore, accurately predicting mine water inflow is critical for drainage system design and water hazard mitigation. Because mine water inflow is governed by the combined influence of multiple hydrogeological factors and thus exhibits pronounced non-linear characteristics, conventional approaches are inadequate in terms of forecasting accuracy and medium- to long-term predictive capability. To address this issue, this study proposes a Chaos-Autoformer-based method for predicting mine water inflow. First, the univariate inflow series is mapped into an m-dimensional phase space by means of phase-space reconstruction from chaos theory, thereby fully preserving its non-linear features; the reconstructed vectors are then used to train and forecast inflow with an improved Chaos-Autoformer model. On top of the original Autoformer architecture, the proposed model incorporates a Chaos-Attention mechanism and a Lyap-Dropout scheme, which enhance sensitivity to small perturbations in initial conditions and complex non-linear propagation paths while improving stability in long-horizon forecasting. In addition, the loss function integrates the maximum Lyapunov exponent error and earth mode decomposition (EMD) indices so as to jointly evaluate dynamical consistency and predictive performance. An empirical analysis based on monitoring data from the Liangshuijing Coal Mine for 2022–2025 demonstrates that the trained model delivers high accuracy and stable performance. Ablation experiments further confirm the significant contribution of the chaos-aware components: when these modules are removed, forecasting accuracy declines to only 76.5%. Using the trained model to predict mine water inflow for the period from June 2024 to June 2025 yields a root mean square error (RMSE) of 30.73 m³/h and a coefficient of determination (R²) of 0.895 against observed data, indicating excellent fitting and predictive capability for medium- to long-term tasks. Extending the forecast to July 2025–November 2027 reveals a pronounced annual cyclical pattern in future mine water inflow, with markedly higher inflow in summer than in winter and an overall slowly declining trend. These findings show that the Chaos-Autoformer can achieve high-precision medium- and long-term predictions of mine water inflow, thereby providing technical support for proactive deployment and refined management of mine water hazard prevention. Full article

Climate change is a global crisis that disproportionately affects vulnerable agrarian communities, exacerbating food insecurity and health risks. This qualitative study explored the relationship between climate change and health in the following two rural sub-counties of Kilifi County, Kenya: Ganze and Magarini. In fall 2023, we conducted 16 focus group discussions with adolescent girls (14–17), young adults (18–30), and older adults (31+). Thematic analysis revealed that climate change adversely affects health through key social determinants, including economic instability, environmental degradation, limited healthcare access, food insecurity, and disrupted education. Participants reported increased food scarcity, disease outbreaks, and reduced access to medical care due to droughts and floods. Economic hardship contributed to harmful survival strategies, including transactional sex and school dropout among adolescent girls. Mental health concerns, such as stress, substance use, and suicidal ideation, were prevalent. These findings highlight the wide-ranging health impacts of climate change in agrarian settings and the urgent need for comprehensive, community-informed interventions. Priorities should include improving nutrition, reproductive and mental health services, infectious disease prevention, and healthcare access. Full article

Introduction: Limited access to pulmonary rehabilitation (PR) has contributed to the rise of telerehabilitation (TPR) for COPD patients. Positive comparable effects are observed in exercise tolerance, quality of life (QoL), and dyspnoea with TPR. However, patient adherence to TPR is an outcome that has not been sufficiently analysed. Objective: To analyse adherence, satisfaction, and quality-of-life improvements in COPD patients following the TPR program to determine whether telerehabilitation is comparable to conventional therapy or usual care. Methods: A systematic search was conducted using four electronic databases, retrieving 392 articles. Two independent researchers selected and evaluated these articles based on predefined eligibility criteria. A third researcher was consulted in the event of disagreements. Results: Primary outcomes: Adherence to PR and/or usual care showed a minimum reported value of 62% and a maximum reported value of 91%, while TPR adherence had the lowest reported value of 21% and the highest reported value of 93.5%. Five articles compared TPR to PR and/or usual care, showing that TPR adherence is higher or similar to other interventions, whereas only one article found lower TPR adherence compared to PR. Secondary outcomes: A higher number of dropouts were reported for PR and usual care compared to TPR. Three publications analysed satisfaction and demonstrated that patients are satisfied across groups. Tertiary outcomes: Comparable improvements in QoL were found for TPR and PR, both being superior to usual care. Conclusions: This systematic review reveals heterogeneity in classifying adherence for pulmonary rehabilitation and telerehabilitation. Adherence classification may be standardised in future studies for consistent analysis. Full article

Overfitting remains a major difficulty in training deep neural networks, especially when attempting to achieve good generalization in complex classification tasks. Standard dropout is often employed to address this issue; however, its uniform random inactivation of neurons typically leads to instability and insufficient performance increases. This paper proposes an upgraded regularization technique merging adaptive sigmoidal dropout with weight amplification, seeking to dynamically adjust neuron deactivation depending on weight statistics, activation patterns, and neuron history. The proposed dropout process uses a sigmoid function driven by a temperature parameter to determine deactivation likelihood and incorporates a “neuron recovery” step to restore important activations. Simultaneously, the method amplifies high-magnitude weights to select crucial traits during learning. The proposed method is tested on CIFAR-10, and CIFAR-100 datasets using four unique CNN architectures, including deep and residual-based models, to evaluate the approach. Results demonstrate that the suggested technique consistently outperforms both standard dropout and baseline models without dropout, yielding higher validation accuracy and lower, more stable validation loss across all datasets. In particular, it demonstrated superior convergence and generalization performance on challenging datasets such as CIFAR-100. These findings demonstrate the potential of the proposed technique to improve model robustness and training efficiency and provide an alternative in complex classification tasks. Full article

The dynamic aperture is an essential concept in circular particle accelerators, providing the extent of the phase space region where particle motion remains stable over multiple turns. The accurate prediction of the dynamic aperture is key to optimising performance in accelerators such as the CERN Large Hadron Collider and is crucial for designing future accelerators like the CERN Future Circular Hadron Collider. Traditional methods for computing the dynamic aperture are computationally demanding and involve extensive numerical simulations with numerous initial phase space conditions. In our recent work, we have devised surrogate models to predict the dynamic aperture boundary both efficiently and accurately. These models have been further refined by incorporating them into a novel active learning framework. This framework enhances performance through continual retraining and intelligent data generation based on informed sampling driven by error estimation. A critical attribute of this framework is the precise estimation of uncertainty in dynamic aperture predictions. In this study, we investigate various machine learning techniques for uncertainty estimation, including Monte Carlo dropout, bootstrap methods, and aleatory uncertainty quantification. We evaluated these approaches to determine the most effective method for reliable uncertainty estimation in dynamic aperture predictions using machine learning techniques. Full article

Background/Objectives: Virtual reality has emerged as a promising intervention for pain management in individuals with cancer. Although its clinical effects have been explored, little is known about participant adherence and dropout behavior. This systematic review and meta-analysis aimed to estimate the pooled dropout rate in randomized controlled trials using virtual reality to treat cancer pain; assess whether dropout differs between groups; and explore potential predictors of attrition. Methods: We conducted a systematic search of PubMed, Web of Science, Scopus, and CINAHL up to April 2025. Eligible studies were randomized trials involving cancer patients or survivors that compared VR interventions for pain management with any non-VR control. Proportion meta-analyses and odds ratio meta-analyses were performed. Heterogeneity was assessed using the I² statistic, and meta-regression was conducted to explore potential predictors of dropout. The JBI appraisal tool was used to assess the methodological quality and GRADE system to determine the certainty of evidence. Results: Six randomized controlled trials were included (n = 569). The pooled dropout rate was 16% (95% CI: 8.2–28.7%). Dropout was slightly lower in VR groups (12.7%) than in controls (21.4%), but the difference was not statistically significant (OR = 0.94; 95% CI: 0.51–1.72; I² = 9%; GRADE: very low). No significant predictors of dropout were identified. Conclusions: VR interventions appear to have acceptable retention rates in oncology settings. The pooled dropout estimate may serve as a reference for sample size calculations. Future trials should improve reporting practices and investigate how VR modality and patient characteristics influence adherence. Full article

Background/Objectives This systematic review and network meta-analysis aimed to determine the most effective therapeutic exercise modality for improving quality of life (QoL) in patients with advanced-stage cancer. Specifically, the study compared the effects of aerobic training, strength training, and combined aerobic and strength training on QoL outcomes. Methods A systematic literature search was conducted in PubMed, Embase, Cochrane Reviews, and the Cochrane Central Register of Controlled Trials up to 24 February 2023. The review adhered to PRISMA guidelines. Included studies were randomized controlled trials (RCTs) involving adult patients with advanced-stage cancers (e.g., pancreatic, colorectal, lung, breast, prostate, gastrointestinal, gynecological, hematological, head and neck, melanoma, or cancers with bone metastases). The primary outcome was post-intervention QoL, while the secondary outcome assessed was the dropout rate across exercise modalities. Results Aerobic training demonstrated the greatest improvement in QoL with a standardized mean difference (SMD) of 0.30 (95% CI: 0.00 to 0.61), followed by strength training (SMD = 0.13; 95% CI: −0.41 to 0.66) and combined training (SMD = 0.07; 95% CI: −0.11 to 0.24). However, none of the interventions showed statistically significant superiority. Dropout rates were comparable across all exercise modalities and control groups, suggesting strong adherence and feasibility of these interventions in advanced cancer populations. Conclusions While all exercise modalities were associated with improved QoL in patients with advanced-stage cancer, no single intervention emerged as significantly superior. Aerobic exercise may offer a slight advantage, although this effect was not statistically significant. These results highlight the importance of individualized exercise prescriptions based on patient preference, functional status, and treatment context. Further research is warranted to identify patient subgroups that may benefit most from specific exercise interventions and to explore QoL subdomains such as fatigue, emotional well-being, and physical functioning. Full article

Background: This study examined treatment outcomes for cannabis use disorder (CUD) among adolescents (12–17 years old) in the United States. Methods: Data from the 2018–2021 Treatment Episode Data Set-Discharges (TEDS-D) included 40,054 adolescents diagnosed with CUD. Descriptive statistics, Chi-square tests, and multivariable logistic regression assessed treatment outcomes and factors associated with treatment completion. Results: Only 36.8% of adolescents completed treatment. The most common reasons for not completing treatment were dropping out (28.4%) and transferring to another facility/program (17.0%). Males and Black non-Hispanic adolescents had lower odds of completing treatment (OR = 0.79, 95%CI: 0.75–0.84), while Hispanic (OR = 1.13, 95%CI: 1.08–1.18), Asian (OR = 1.56, 95%CI: 1.3–1.86) and Native Hawaiian/Pacific Islander adolescents (OR = 2.31, 95%CI: 2.04–2.61) had higher odds of completion compared to their White counterparts. Independent living arrangements, homelessness, arrests in the past 30 days and younger age (<15 years old) decreased the likelihood of treatment completion. Adolescents with co-occurring mental health and substance use disorders also had lower completion rates (OR = 0.79, 95%CI: 0.77–0.86). Referral from schools/employers and treatment settings were associated with a higher success, particularly with stays of 4–6 months and 7–12 months. Conclusion: This study highlights the need for targeted CUD treatment programs that support at-risk adolescents, especially those experiencing homelessness or facing legal issues. High dropout and transition rates suggest a need for continuity of care and program integration between facilities. Strengthening coordination among public health officials, community organizations, and stakeholders is essential to developing culturally responsive treatment interventions that address social determinants of health, substance use, and mental health in this vulnerable population. Full article

Sewer pipelines are essential urban infrastructure that play a key role in sanitation and disaster prevention. Regular condition assessments are necessary to detect defects early and determine optimal maintenance timing. However, traditional visual inspection using closed-circuit television (CCTV) footage is time-consuming, labor-intensive, and dependent on subjective human judgment. To address these limitations, this study develops a convolutional neural network (CNN)-based sewer defect classification model and analyzes how data quality—such as mislabeled or redundant images—affects model accuracy. A large-scale public dataset of approximately 470,000 sewer images was used for training. The model was designed to classify non-defect and three major defect categories. Based on the ResNet50 architecture, the model incorporated dropout and L2 regularization to prevent overfitting. Experimental results showed the highest accuracy of 92.75% at a dropout rate of 0.2 and a regularization coefficient of 0.01. Further analysis revealed that mislabeled, redundant, or obscured images within the dataset negatively impacted model performance. Additional experiments quantified the impact of data quality on accuracy, emphasizing the importance of proper dataset curation. This study provides practical insights into optimizing data-driven approaches for automated sewer defect detection and high-performance model development. Full article

Objectives: This study aims to assess the clinical efficacy of micro-fragmented adipose tissue (MFAT) therapy over three years in patients with KOA and to determine whether short-term improvements at three months can forecast long-term outcomes. Methods: A retrospective, observational study was conducted on 335 patients diagnosed with KOA who received a single MFAT injection. The patients were followed up at 3 months, 6 months, 1 year, 2 years, and 3 years, with assessments using the Visual Analog Scale (VAS), Oxford Knee Score (OKS), Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), and Knee Injury and Osteoarthritis Outcome Score (KOOS). Statistical analysis was performed to assess the differences in preoperative and postoperative scores (VAS, OKS, WOMAC, KOOS) to evaluate the predictive role of 3-month score changes on long-term clinical outcomes. Results: All measured scores (VAS, OKS, WOMAC, KOOS) showed significant improvement at 3 months, with sustained improvements through 3 years (p < 0.001). Early score changes at 3 months were significantly associated with improved clinical outcomes at 1, 2, and 3 years (p < 0.05). Logistic regression confirmed early post-treatment improvements as independent predictors of long-term benefit, except for the VAS score at 3 years (p = 0.098). A comparative analysis between completers and dropouts showed no baseline differences; however, significant outcome differences emerged at later follow-up points. Due to insufficient data at the 3-year mark among dropouts, statistical comparisons were not possible for that time point. Conclusions: MFAT treatment was associated with consistent symptomatic improvement in patients with KOA, and early clinical response at 3 months served as a reliable predictor of long-term pain and function outcomes. While this study focused on patient-reported symptom relief and not structural regeneration, the results support MFAT as a minimally invasive option for symptom management. Early post-treatment response may serve as a useful tool for clinicians to predict long-term therapeutic success and personalize treatment strategies for KOA patients. Full article

Background and Objectives: There have been a limited number of randomized controlled trials (RCTs) comparing pharmacopuncture therapy (PPT) and physical therapy (PT) for chronic knee pain. In this study, we assess the feasibility, safety, and preliminary effectiveness of PPT compared to PT in patients with chronic knee pain. Materials and Methods: This pilot study was designed as a two-arm, parallel RCT. Patients were recruited through in-hospital advertisements. Forty patients aged 19 to 70 with knee pain with a numeric rating scale (NRS) score of 5, persisting for >3 months, were randomized into the PPT or PT group. The type of PT solution or PT method was not determined in advance, leaving it to the clinician’s judgment. Treatment was administered twice weekly for 3 weeks with a 6-week follow-up. The primary outcome was the NRS score for knee pain, whereas the secondary outcomes were the visual analog scale (VAS), knee range of motion, Korean Western Ontario and McMaster (K-WOMAC), Patient Global Impression of Change, and five-level EuroQol five-dimension scores. Additionally, adherence, acceptability, dropout rate, and adverse events were measured to assess the feasibility of a follow-up main study. The protocol was registered at ClinicalTrials.gov (NCT06505681). Results: The PPT group showed significantly superior improvement compared with the PT group in the NRS (difference = −2.05, 95% confidence interval [CI]: −2.76 to −1.34), VAS (difference = −21.58, 95% CI: −29.42 to −13.74), and K-WOMAC scores (difference = −13.17, 95% CI: −21.67 to −4.67). Of the 55 patients who initially expressed interest in participation, 8 declined after receiving detailed information about this study. Among the forty enrolled participants, one patient in the PPT group dropped out, and one missed a single treatment session. Apart from these cases, all participants completed the assigned treatments and follow-up assessments, demonstrating high adherence. No serious adverse events were reported. Conclusions: PPT demonstrated excellent effectiveness in pain relief and functional improvement in these patients. Full article

Educational data mining (EDM) is instrumental in the early detection of students at risk of academic underperformance, enabling timely and targeted interventions. Given that many undergraduate students face challenges leading to high failure and dropout rates, utilizing EDM to analyze student data becomes crucial. By predicting academic success and identifying at-risk individuals, EDM provides a data-driven approach to enhance student performance. However, accurately predicting student performance is challenging, as it depends on multiple factors, including academic history, behavioral patterns, and health-related metrics. This study aims to bridge this gap by proposing a deep learning model to predict student academic performance with greater accuracy. The approach combines a convolutional neural network (CNN) and a bidirectional gated recurrent unit (BiGRU) network to enhance predictive capabilities. To improve the model’s performance, we address key data preprocessing challenges, including handling missing data, addressing class imbalance, and selecting relevant features. Additionally, we incorporate optimization techniques to fine-tune hyperparameters to determine the best model architecture. Using key performance metrics such as accuracy, precision, recall, and F-score, our experimental results show that our proposed model achieves improved prediction accuracy of 97.48%, 90.90%, and 95.97% across the three datasets. Full article

Background/Objectives: Exercise has gained attention as a potentially beneficial non-pharmacological intervention, but whether this type of intervention presents a higher dropout rate compared to other interventions is still unknown. This systematic review, with a meta-analysis of randomized controlled trials, aims to determine whether exercise or comparators present lower or higher attrition in patients with migraine. Methods: A search was conducted in PubMed, Scopus, Web of Science, and Cochrane Library until March 2025. The methodological quality was evaluated using the JBI scale for randomized trials. Proportion meta-analysis calculated the dropout rate. Results: Odds ratio meta-analysis under 1 indicated lower attrition in experimental participants. Subgroup meta-analyses sorted by type of exercise, control, and migraine were conducted to explore variability in results based on the mentioned moderators. The overall pooled dropout rate was 6.7%, 11.6% for the exercise groups, and 10.1% for the comparators. No statistical difference was found between groups of studies, type of migraine, type of exercise, and type of comparator (p ≥ 0.05). Only the odds ratio results for migraine with auras showed a lower pooled dropout rate in favor of control participants, OR = 1.18. Conclusions: Although there is no statistically significant difference, the meta-analysis of proportions shows a higher loss rate in exercise-based interventions. However, the high heterogeneity found in the included studies prevents us from drawing firm conclusions. Furthermore, adequate adherence to the CONSORT guidelines in reporting losses and their reasons could help design appropriate retention strategies for studies and interventions based on exercise in patients with migraines. Full article