Browse Articles

Multi-task reinforcement learning and explainable AI-Driven platform for personalized planning and clinical decision support in orthodontic-orthognathic treatment

Abstract This study presents a novel clinical decision support platform for orthodontic-orthognathic treatment that integrates multi-task reinforcement learning with explainable artificial intelligence. The platform addresses the challenges of personalized treatment planning in complex dentofacial deformities by formulating treatment as a sequential decision-making process optimizing multiple clinical objectives simultaneously. We developed a comprehensive framework comprising: (1) a multi-task reinforcement learning core with specialized state-action representations for craniofacial structures; (2) complementary explainable AI components that render complex model decisions interpretable within clinical contexts; and (3) an interactive interface facilitating collaborative human-AI decision-making. Experimental validation with 347 retrospectively analyzed cases demonstrated significant improvements in treatment plan quality (19.9%), decision efficiency (73.9% time reduction), and prediction accuracy (92.7%) compared to conventional methods. Clinical evaluation by multidisciplinary specialists confirmed the system’s practical utility, with particularly strong performance in complex cases featuring multiple dentofacial abnormalities. The proposed platform represents a significant advancement toward evidence-driven, personalized treatment planning in orthodontic-orthognathic therapy while maintaining clinical interpretability and expert oversight.

Exploring the variation in muscle response testing accuracy through repeatability and reproducibility

Research Objectives To explore the variation in mean muscle response testing (MRT) accuracy and whether this variation can be attributable to participant characteristics. Methods A prospective study of diagnostic test accuracy was carried out in a round-robin format. Sixteen practitioners tested each of 7 test patients (TPs) using 20 MRTs broken into 2 blocks of 10 which alternated with 2 blocks of 10 intuitive guessing. Mean MRT accuracies (as overall percent correct) were calculated for each unique pair. Reproducibility and repeatability were assessed using analyses of variance (ANOVA) and scatter and Bland-Altman plots. Results The mean MRT accuracy (95% CI) was 0.616 (0.578–0.654), which was significantly different from both the mean intuitive guessing accuracy, 0.507 (95% CI 0.484–0.530; p<0.01) and chance (p<0.01). Visual inspection of scatterplots of mean MRT accuracies by practitioner and by TP suggest large variances among both subsets, and regression analysis revealed that MRT accuracy could not be predicted by TP (r = ‒0.14; p = 0.19), nor by Practitioner (r = 0.01; p = 0.90). A significant effect imposed by both practitioners and TPs individually and together was found at the p<0.05 level; however, together they account for only 57.0% of the variance, with 43.0% of the variance unexplained by this model. From a statistical perspective, Bland-Altman Plots of mean MRT accuracy by practitioner do show adequate repeatability since all scores fell within 2 SDs of the mean; however, the wide range of scores also suggests insufficient repeatability from a clinical perspective. Finally, ANOVA demonstrated that an insignificant amount of variance could be explained by block [F(1,21) = 0.02, p = 0.90].

Interaction of genetic risk score (GRS) and Plant-Based diet on atherogenic factors and body fat distribution indices among women with overweight and obesity: a cross-sectional study

Abstract The association between plant-based diets, obesity, cardiovascular disease (CVD), and genetic predisposition is still not fully understood. This study explored how plant-based diets interact with genetic susceptibility to atherosclerosis and body fat in 377 Iranian women aged 18 to 48 who were overweight or obese. Using a validated 147-item food frequency questionnaire (FFQ), we established three plant-based diet indices: the Plant-Based Diet Index (PDI), the Healthy Plant-Based Diet Index (hPDI), and the Unhealthy Plant-Based Diet Index (uPDI). We calculated a Genetic Risk Score (GRS) based on three body mass index (BMI)-related single nucleotide polymorphisms (SNPs) and analyzed its interaction with the PDI. Results showed that tertile 2 of the PDI had a significant negative interaction with moderate and high-risk alleles on key atherogenic factors, including the atherogenic index of plasma (AIP), triglyceride glucose (TyG), lipid accumulation product (LAP), and visceral adiposity index (VAI) (P < 0.05). A borderline negative interaction between tertile 2 of the hPDI and moderate risk alleles regarding the Body Shape Index (ABSI) was also observed (P = 0.05). Conversely, tertile 2 of the uPDI had a significant positive interaction with moderate risk alleles related to both the Castelli Risk Index I (CRI.I) and II (CRI.II) (P = 0.03). This study suggests that interactions between genetic susceptibility and plant-based diet indices are linked to atherogenic factors and body composition.

Utilizing statistical analysis for motion imagination classification in brain-computer interface systems

In this study, we introduce a novel Field-Agnostic Riemannian-Kernel Alignment (FARKA) method to advance the classification of motion imagination in Brain-Computer Interface (BCI) systems. BCI systems enable direct control of external devices through brain activity, bypassing peripheral nerves and muscles. Among various BCI technologies, electroencephalography (EEG) based on non-intrusive cortical potential signals stands out due to its high temporal resolution and non-invasive nature. EEG-based BCI technology encodes human brain intentions into cortical potentials, which are recorded and decoded into control commands. This technology is crucial for applications in motion rehabilitation, training optimization, and motion control. The proposed FARKA method combines Riemannian Alignment for sample alignment, Riemannian Tangent Space for spatial representation extraction, and Knowledge Kernel Adaptation to learn field-agnostic kernel matrices. Our approach addresses the limitations of current methods by enhancing classification performance and efficiency in inter-individual MI tasks. Experimental results on three public EEG datasets demonstrate the superior performance of FARKA compared to existing methods.

Machine learning assisted adjustment boosts efficiency of exact inference in randomized controlled trials

Associations between ambient particulate matter exposure and the prevalence of arthritis: Findings from the China Health and Retirement Longitudinal Study

Background Arthritis is a leading cause of global disability, but its etiology is complex and has not yet been fully understood. Recent studies suggest that air pollution is potentially linked to arthritis onset, although current research predominantly focuses on the effects of particulate matter (PM 2.5 ). There is a lack of comprehensive analysis regarding the influence of various particulate pollutants and their key constituents on arthritis prevalence. This research used China Health and Retirement Longitudinal Study (CHARLS) data and explored the associations between arthritis prevalence and different particulate matters (PM 2.5 , PM 10 , and PM 1 , ≤ 2.5 micrometers, ≤ 10 micrometers, and ≤ 1 micrometer in diameter, respectively), as well as ammonium (NH 4 ) and nitrate (NO 3 ), with gaseous pollutants [such as sulfate (SO 4 ) and ozone (O 3 )] as a reference. Methods This study was conducted based on the 2015 CHARLS cross-sectional data, and 3,802 participants were included. The levels of air pollution exposure were estimated using a spatial-temporal extreme random forest model, integrating ground monitoring, remote sensing data, and model simulations, encompassing PM 1 , PM 2.5, PM 10 , NH 4 , NO 3 , O 3 , and SO 4 concentrations. The association of air pollution with arthritis prevalence was assessed utilizing generalized linear models (GLM), while adjusting for various confounding variables. Results Long-term exposure to PM₁, PM₂.₅, PM₁₀, NH₄ ⁺ , and NO₃ ⁻ was positively associated with self-reported arthritis prevalence. Specifically, each interquartile range (IQR) increase in PM₁ corresponded to 4.4% higher odds of arthritis (odds ratio [OR] per IQR: 1.044; 95% confidence interval [CI]: 1.011–1.070), indicating a modest association. Subsequent ORs per IQR for PM₂.₅ and PM₁₀ were 1.019 (95% CI: 1.004–1.036) and 1.012 (95% CI: 1.002–1.021), respectively, reflecting similar but smaller positive associations, whereas NH₄⁺ and NO₃ ⁻ showed moderate associations with ORs per IQR of 1.143 (95% CI: 1.017–1.285) and 1.082 (95% CI: 1.011–1.159), respectively. These findings were robust to sensitivity analyses. Conclusion To our knowledge, this is the first study to identify a significant association between long-term exposure to PM₁, PM₂.₅, PM₁₀ and the secondary inorganic aerosol constituents NH₄ and NO₃, and the prevalence of self-reported arthritis in middle-aged and older adults. However, owing to its cross-sectional design, the absence of subtype differentiation and reliance on self-reported diagnoses, these findings may be influenced by reverse causation and measurement error, and should therefore be interpreted with caution.

Four methods to identify sensitive periods of infant weight gain associated with adolescent BMI in a Chinese birth cohort

Simulation of emitter discharge along drip laterals under drip fertigation system using artificial neural network

Simulation of emitter discharge under a drip fertigation system is important for capturing the variation in water and nutrient distribution to crops. This is important for an effective design and irrigation management for agricultural crops. Moreover, the field discharge measurements are laborious and time-consuming, hence the need for the development of a representative model. The application of artificial neural network to simulate drip emitter along drip laterals is new in the field of flow measurement under drip irrigation. The purpose of this study is to predict the emitter discharge along drip laterals using artificial neural network (ANN) and evaluate the performance of the model. The input parameters fed into the ANN include; pipe length away from the fertigation source, elevation heads and distance of emitter point along the laterals. The field measured discharge was considered as the output. Evaluation parameters considered for the designed drip fertigation system indicated high efficiency, in the range between 81 and 98%. Interaction effects were observed between the pipe length and elevation head on the uniformity coefficient (CU) and emitter discharge. When all data were simulated, the ANN model simulated the emitter discharge accurately and precisely along the drip laterals, with R 2 value ranging between 0.81 and 0.89, while the normalized root mean square error (NRMSE) was mostly below 20%, thus indicating a good prediction. The mean absolute error ranged between 0.034 and 0.048. Therefore, the ANN model was efficient for capturing the variation in emitter discharge well under the drip fertigation system.

Structural health monitoring and evaluation method for an immersed tunnel based on deep learning

Significance of the cribriform pattern in predicting the prognosis of lung adenocarcinoma patients: A systematic review and meta-analysis

Background A number of studies have shown that various histological subtypes of lung adenocarcinoma have different clinical prognoses, but the cribriform pattern, as a unique histological subtype, plays an important role in the prognosis of patients with lung adenocarcinoma. Objective In this meta-analysis, we evaluated the role of the cribriform pattern in the overall survival of patients with lung adenocarcinoma, which may provide valuable information for the treatment of patients with lung adenocarcinoma. This also provides an important basis for dividing the cribriform pattern into a new histological subtype and classifying it. Methods We searched the literature from the PubMed, Embase, Cochrane and Web of Science online databases; extracted the data and characteristics of each study; and extracted and calculated hazard ratios (HRs) with 95% confidence intervals (CIs) to evaluate the impact of the cribriform pattern on the prognosis of patients with lung adenocarcinoma. Results A total of 10 articles were included in the study according to the preset criteria, with a total of 5487 research subjects. The hazard ratio for the relationship between the cribriform pattern and the overall survival rate of patients with lung adenocarcinoma was 2.05 (95% CI: 1.76–2.39). The prognosis of patients with positive spread through air spaces in the cribriform pattern was significantly worse than that of patients with negative spread through air spaces in the cribriform pattern, with a hazard ratio of 2.58 (95% CI: 1.84–3.62). There was a significant difference in prognosis between patients with the cribriform pattern and those with low-grade and intermediate-grade lung adenocarcinoma, but there was no significant difference in prognosis between patients with the cribriform pattern and those with high-grade lung adenocarcinoma, with hazard ratios of 2.12 (95% CI: 1.12–4.00), 7.70 (95% CI: 2.15–36.20) and 0.96 (95% CI: 0.54–2.40), respectively. Therefore, the cribriform pattern should be used as a histological subtype of high-grade tumors, thus influencing the postoperative prognosis of patients with lung adenocarcinoma. Conclusion The presence of the cribriform pattern is an independent risk factor for postoperative overall survival in patients with lung adenocarcinoma. The cribriform pattern should be considered a new histological subtype of lung adenocarcinoma and classified with solid carcinoma and micropapillary adenocarcinoma as high-grade tumors.

Social anxiety and gratitude among children left behind due to migration in western rural China

Expression of Concern: Facemask wearing to prevent COVID-19 transmission and associated factors among taxi drivers in Dessie City and Kombolcha Town, Ethiopia

Smartwatch-based wrist tremor assessment in neurosurgical simulator training

Urban-rural disparities in skilled birth attendance among women in Ethiopia: Multivariate decomposition analysis

Introduction Skilled birth attendants play an important role in reducing maternal mortality. Although Ethiopia has shown a remarkable reduction in maternal mortality, maternal health service utilization, such as skilled birth attendance, remains low. Thus, this study aims to assess the urban-rural disparity in skilled birth attendance in Ethiopia using the 2019 Ethiopian mini demographic health survey. Methods and materials The study was based on data obtained from demographic and health surveys in Ethiopia. A total weighted sample of 5,527 women who gave birth within 5 years preceding the survey was included. The result of descriptive statistics was reported using the frequency, percentages, graphs, and tables. A multivariate decomposition analysis was used to identify factors contributing to the disparity of skilled birth attendance across residence. Statistical significance was defined at a 95% confidence interval with a p-value of less than 0.05. Result Skilled birth attendance utilization among women in Ethiopia was 49.8% (95% CI: 48.5–51.1). The disparity in skilled birth attendance coverage between urban and rural areas was significantly high (Urban coverage was 72.1% and rural coverage was 42.5%). Endowment coefficients (women’s characteristics) explained 88% of the urban-rural disparity in the magnitude of skilled birth attendance. Women with secondary and above educational status, four or more antenatal care visits, households with televisions and radio, women in the richest wealth index and women with five or more parity were the determinants that explained the urban-rural disparity in skilled birth attendance. Conclusion and recommendations There was a significant disparity in skilled birth attendance utilization between urban and rural areas. Factors like maternal education, wealth status, antenatal care visits, and media access explained the disparity. To attain equitable progress towards universal coverage of SBA, special efforts and resources should be targeted towards rural women. Initiatives aimed at enhancing access to health services and health care consultations for the rural community are also recommended.

Deep learning-based allergic rhinitis diagnosis using nasal endoscopy images

Addressing the barriers to peritoneal dialysis—Visual appeal matters

Peritoneal dialysis (PD) is an effective renal replacement strategy for patients with end-stage renal disease utilizing the peritoneum as the filter and PD catheter as access. A survey of PD patients showed that some felt uncomfortable with the length of current catheters and would be interested to explore newer, shorter designs. We redesigned the transfer set and external portion of the catheter, addressing this barrier as a part of our multi-institutional design project led by a nephrologist from the University of Arkansas for Medical Sciences (UAMS) and an engineering design team from the biomedical engineering department at the University of Arkansas. Multiple designs were considered, including spiral, retractable, and collapsible bulbs, with an accordion-style mechanism being selected for prototyping. We created several prototypes, first by 3D printing as well as by silicone casting. Computational fluid analysis showed the design to be fully capable of delivering clinically relevant flows. The final design of our PD transfer set has a flexible accordion section that is 7 cm when extended and collapses to substantially shorten this length. We propose that the design can also be extended to the extra-abdominal section of the PD catheter.

The impact of adding dual and triple combinations of quicklime and plastic wastes and palm fibers on the California bearing ratio of fine sand

Abstract Fine sand, widely distributed across arid and semi-arid regions, presents challenges due to its low bearing capacity and susceptibility to deformation. This study investigates the enhancement of the California Bearing Ratio (CBR) value of fine sand through the incorporation of palm fibers, plastic waste, and quicklime. This study investigates the enhancement of the CBR of fine sand using palm fibers, plastic waste, and quicklime. Through two experimental phases, optimal dosages were determined as 5.0% quicklime, 0.75% plastic waste, and 1.0% palm fiber. Single additive treatments yielded CBR improvements of 195%, 125%, and 275%, respectively. Combinations revealed that mixing quicklime with palm fibers decreased enhancement efficiency due to chemical incompatibility. Notably, palm fiber–plastic waste mixtures proved more sustainable. This research offers a cost-effective, eco-friendly solution for improving subgrade conditions with clear quantitative outcomes. The findings underscore the potential of recycling plastic waste by mixing with palm fibers for sustainable improvement of fine sand properties. By reducing plastic pollution and encouraging circular resource usage, the combination of natural palm fibers and recovered plastic waste enhances soil performance and promotes environmental sustainability. Furthermore, this approach is more affordable than traditional soil stabilizing methods, especially in areas where local resources are easily accessible.

Expression of Concern: Facemask-wearing behavior to prevent COVID-19 and associated factors among public and private bank workers in Ethiopia

Healthcare professionals’ knowledge, attitudes, and practices towards predictive diagnosis of early neurological deterioration

Global, regional, and national burden of near vision loss in children and adolescents under 20 years from 1990–2021 and prediction to 2060: A cross-sectional study based on the global burden of disease study 2021

Near vision loss (NVL) has become a significant global public health concern, particularly among children and adolescents under 20 years, who face increasing academic demands and prolonged screen exposure. The COVID-19 pandemic, characterized by excessive screen time and reduced outdoor activities, has likely exacerbated this trend. This study analyzes the global, regional, and national burden of NVL from 1990 to 2021 and projects future trends up to 2060 using data from the Global Burden of Disease (GBD) Study 2021. Prevalence and Disability-Adjusted Life Years (DALYs) associated with NVL were assessed across different socioeconomic levels, and future trends were forecasted using the Bayesian Age-Period-Cohort (BAPC) model. Results indicate a significant increase in NVL cases, rising to 31.7 million in 2021, with projections reaching 33 million by 2060. A strong negative correlation was observed between the Social Development Index (SDI) and NVL burden, with Africa exhibiting the highest prevalence and Australasia the lowest. Notably, NVL burden in higher SDI regions rebounded post-COVID-19, reversing previous declining trends. Across all age groups, NVL prevalence continues to rise, with females consistently exhibiting higher rates than males. These findings underscore the urgent need for targeted public health policies and resource allocation strategies to mitigate the rising burden of NVL among children and adolescents, particularly in lower SDI regions. Addressing modifiable risk factors, promoting early interventions, and integrating vision care into public health frameworks will be crucial in managing this growing health crisis.