Bioengineering最新文献

Due to the damage happening in the nervous system, neuropathic pain occurs and it affects the quality of life of the patient to a great extent. Therefore, some clinical evaluations are required to assess the diagnostic outcomes precisely. A lot of information about the activities of the brain is provided by Electroencephalography (EEG) signals and neuropathic pain can be assessed and classified with the aid of EEG and machine learning. In this work, two approaches are proposed in terms of efficient blended models for the classification of neuropathic pain through EEG signals. In the first blended model, once the features are extracted using Discrete Wavelet Transform (DWT), statistical features, and Fuzzy C-Means (FCM) clustering techniques, the features are selected using Grey Wolf Optimization (GWO), Feature Correlation Clustering Technique (FCCT), F-test, and Bayesian Optimization Algorithm (BOA) and it is classified with the help of three hybrid classification models like Spider Monkey Optimization-based Gradient Boosting Machine (SMO-GBM) classifier, hybrid deep kernel learning with Support Vector Machine (DKL-SVM) classifier, and CatBoost classifier. In the second blended model, once the features are extracted, the features are selected using Hybrid Feature Selection-Majority Voting System (HFS-MVS), Hybrid Salp Swarm Optimization-Particle Swarm Optimization (SSO-PSO), Pearson Correlation Coefficient (PCC), and Mutual Information (MI) and it is classified with the help of three hybrid classification models like Partial Least Squares (PLS) variant classification models combined with Kernel-based SVM, ensemble classification model with soft voting strategy, and Extreme Gradient Boosting (XGBoost) classifier. The proposed blended models are evaluated on a publicly available dataset and the best results are shown when the FCM features are selected with SSO-PSO feature selection technique and classified with Polynomial Kernel-based PLS-SVM Classifier, reporting a high classification accuracy of 92.68% in this work.

Error in Author List [...].

Background: Dentofacial anomalies, including malocclusion, emerge from the interplay of genetic, clinical, and environmental determinants. Understanding the factors associated with these anomalies is crucial at the primary care level. Our study aimed to determine the possible associated factors with dentofacial anomalies in patients attended at the primary care level. Methods: A multivariate logistic regression model was applied to a primary care population, with the presence of dentofacial anomalies as the dependent variable. Independent variables included age and selected clinical conditions of dental and neurological origin. Results: Age was inversely associated with dentofacial anomalies (OR = 0.991; 95% CI 0.985-0.998; p = 0.013). Significant clinical factors included vertigo (OR = 2.59; 95% CI 1.42-4.71; p = 0.002), hearing loss (OR = 4.34; 95% CI 2.44-7.72; p < 0.001), trigeminal neuralgia (OR = 8.54; 95% CI 3.22-22.67; p < 0.001), Bell's palsy (OR = 9.19; 95% CI 4.01-21.04; p < 0.001), caries limited to enamel (OR = 17.92; 95% CI 12.99-24.71; p < 0.001), and acute gingivitis (OR = 10.64; 95% CI 5.61-20.20; p < 0.001). Conclusions: Both oral and neurological conditions showed strong associations with dentofacial anomalies. The model identified key factors that may facilitate early detection and guide the development of targeted preventive strategies in oral health practice and policy, supporting the integration of multidisciplinary approaches to patient care.

Meniscal repair has become the preferred treatment for many meniscal tears. As a result, multiple arthroscopic techniques have evolved, including the all-inside (AI) and inside-out (IO) approaches, which have been widely studied in the current literature. The present article highlights key limitations in studies reporting long-term outcomes (≥5 years), notably the heterogeneity of failure definitions and the lack of subgroup stratification by clinically relevant factors such as age, concomitant anterior cruciate ligament reconstruction (ACLR), and meniscal side (medial vs. lateral). To date, no clear superiority of the AI over the IO approach has been established. Redefining failure through multidimensional approaches that integrate structural, clinical, and patient-reported assessments will be crucial to ensure a consistent and patient-centered evaluation of repair success. Further research with robust subgroup analyses is needed to determine whether one technique confers superior long-term results in specific patient populations.

Knee osteoarthritis (OA) affects around 37% of U.S. adults over 60, with over 25% experience depressive symptoms (DSs), linked to worse pain and outcomes. Yet their impact on analgesic use and recovery remains unclear. This study aimed to assess if DSs influence analgesic use and functional outcomes in knee OA. Using data from the Osteoarthritis Initiative (n = 3680), we used a Machine Learning (ML)-based Gradient Boosting Machine (GBM) model to perform propensity score matching, matching patients with knee OA and DSs (n = 487) to those without DSs (n = 487). Outcomes at baseline, 1 and 2 years included analgesic use, function (WOMAC), quality of life (KOOS-QoL), and physical health (SF-12 PCS). Regression and timepoint models compared follow-up with baseline. DSs alone were not associated with greater opioid use up to Year 2 (OR = 0.89, 95% CI: 0.45-1.73; p = 0.73). Among patients with DSs, SF-12 PCS improvement was less likely at Year 1, while decline was more likely up to Year 2. DSs in OA were linked to poorer physical health, but often greater functional gains than those in OA without DSs, with no difference in opioid use. These findings highlight the need for multidisciplinary strategies, addressing both pain and psychosocial wellbeing.

This study evaluated the feasibility, image quality, and referral accuracy of a patient-operated optical coherence tomography (OCT) device (SightSync) compared with technician-acquired Heidelberg OCT. This study was conducted in a Veterans Affairs retina clinic (Cleveland, Ohio), resulting in a predominantly male (98%) study population representative of the local veteran demographics. One hundred patients attempted self-administered OCT imaging after brief instruction, yielding 118 successful scans (59% of eyes) with no significant association between scan success and age, visual acuity, or diagnosis. Quantitative analysis of 142 paired images showed that SightSync produced interpretable scans with comparable sharpness to Heidelberg OCT, though signal- and intensity-based metrics (signal-to-noise ratio; SNR, contrast-to-noise ratio; CNR, entropy, pixel intensity; p90) were lower, consistent with hardware differences between a compact patient-operated prototype and a clinical-grade system. Among 121 high-quality SightSync scans, referral decisions demonstrated strong agreement with Heidelberg OCT, with a sensitivity of 83.9%, specificity of 75.6%, and a negative predictive value of 93.2%, indicating reliable exclusion of clinically significant pathology. These findings demonstrate that patients can independently acquire clinically interpretable OCT images and that SightSync provides safe, conservative triage performance-supporting its potential as a scalable community-based retinal imaging solution-while a review of unsuccessful scans has identified prototype modifications expected to further improve device feasibility.

The urgent need for effective food waste management, coupled with the scarcity of carbon sources for sewage treatment, highlights the potential of producing carbon sources from food waste as a mutually beneficial solution. This study investigated the production of carbon sources through anaerobic fermentation using the liquid phase of food waste three-phase separation. Compared with previous studies using raw food waste or mixed substrates, the liquid phase derived from three-phase separation is richer in soluble organic matter and has been pre-heated (80 °C), which facilitates subsequent fermentation and offers easier integration into existing food waste treatment plants. A series of lab-scale batch fermentation experiments were carried out at different temperatures, including ambient, mesophilic, and thermophilic conditions, as well as varying initial pH levels (uncontrolled, neutral, and alkaline). The experimental results indicated that optimal production parameters involve a 4-day mesophilic fermentation at 35 °C with an initial alkaline pH, which increased the total VFAs yield by 252.5% to 40.26 g/L and raised the acetic acid fraction to 45.5% of total VFAs. Under these conditions, there was an observed increase in the relative abundance of acidogenic bacteria and a decrease in that of methanogen archaea. Furthermore, the denitrification performance of the produced carbon source was evaluated in short-term tests, and near-complete nitrate removal was achieved within approximately 2 h. These findings suggest the fermented liquid phase of food waste is a promising partial substitute for conventional external carbon sources.

Background: Immersive virtual reality (VR) has emerged as a promising tool to enhance neuroplasticity, motivation, and engagement during post-stroke motor rehabilitation. However, evidence on its feasibility and data-driven integration into clinical practice remains limited.

Objective: This pilot study aimed to evaluate the feasibility, usability, and short-term motor outcomes of an immersive VR-assisted rehabilitation program using the Travee-VR system.

Methods: Fourteen adults with post-stroke upper-limb paresis completed a 10-day hybrid rehabilitation program combining conventional therapy with immersive VR sessions. Feasibility and tolerability were assessed through adherence, adverse events, the System Usability Scale (SUS), and the Simulator Sickness Questionnaire (SSQ). Motor outcomes included active and passive range of motion (AROM, PROM) and a derived GAP index (PROM-AROM). Correlations between clinical changes and in-game performance metrics were explored to identify potential digital performance metrics of recovery.

Results: All participants completed the program without adverse events. Usability was rated as high (mean SUS = 79 ± 11.3), and cybersickness remained mild (SSQ < 40). Significant improvements were observed in shoulder abduction (+7.3°, p < 0.01) and elbow flexion (+5.8°, p < 0.05), with moderate-to-large effect sizes. Performance gains in the Fire and Fruits games correlated with clinical improvement in shoulder AROM (ρ = 0.45, p = 0.041). Cluster analysis identified distinct responder profiles, reflecting individual variability in neuroplastic adaptation.

Conclusions: The Travee-VR system proved feasible, well tolerated, and associated with measurable short-term improvements in upper-limb function. By linking clinical outcomes with real-time kinematic data, this study supports the role of immersive, feedback-driven VR as a catalyst for data-informed neuroplastic recovery. These results lay the groundwork for adaptive, clinic-to-home rehabilitation models integrating clinical and exploratory digital performance metrics.

Fatigue is a multifactorial phenomenon affecting both physical and psychological performance, particularly in high-stress occupations. Although wearable sensors enable continuous monitoring, conventional machine-learning (ML) models can produce unstable, weakly calibrated, and opaque predictions in real-world settings. To improve reliability and interpretability, we developed a selective Retrieval-Augmented Generation (RAG)-enhanced hybrid ML-LLM framework that integrates the efficiency of ML with the reasoning capability of large language models (LLMs). Using wearable and ecological momentary assessment data from 297 emergency responders (9543 seven-day windows), logistic regression, XGBoost, and LSTM models were trained to classify fatigue levels dichotomized by the median of daily tiredness scores. The LLM was selectively activated only for borderline ML outputs (0.45 ≤ p ≤ 0.55), using symbolic rules and retrieved analog examples. In the uncertainty region, performance improved from 0.556/0.684/0.635/0.659 to 0.617/0.703/0.748/0.725 (accuracy/precision/recall/F1). On the full test set, performance similarly improved from 0.707/0.739/0.918/0.819 to 0.718/0.741/0.937/0.827, with gains confirmed by McNemar's paired comparison test (p < 0.05). SHAP-based ML interpretation and LLM reasoning analyses independently identified short-term sleep duration and heart-rate variability as dominant predictors, providing transparent explanations for model behavior. This framework enhances classification robustness, interpretability, and efficiency, offering a scalable solution for real-world fatigue monitoring.

Background: Sitting and standing with conventional hip-knee-ankle-foot (HKAF) prostheses are demanding tasks for hip disarticulation (HD) amputees due to the passive nature of current prosthetic hip joints that cannot assist with moment generation. This study developed a sitting and standing control strategy for a motorized hip joint and evaluated whether providing active assistance reduces the intact side demand of these activities. Methods: A dedicated control strategy was developed and implemented for a motorized hip prosthesis (Power Hip) compatible with existing prosthetic knees, feet, and sockets. One HD participant was trained to perform sitting and standing tasks using the Power Hip. Its performance was compared with the participant's prescribed passive HKAF prosthesis through measurements of ground reaction forces (GRFs), joint moments, and activity durations. GRFs were collected using force plates, kinematics were captured via Theia3D markerless motion capture, and joint moments were computed in Visual3D. Results: The Power Hip enabled more symmetric limb loading and faster stand-to-sit transitions (1.22 ± 0.08 s vs. 2.62 ± 0.41 s), while slightly prolonging sit-to-stand (1.69 ± 0.49 s vs. 1.22 ± 0.40 s) compared to the passive HKAF. The participant exhibited reduced intact-side loading impulses during stand-to-sit (4.97 ± 0.78 N∙s/kg vs. 15.06 ± 2.90 N∙s/kg) and decreased reliance on upper-limb support. Hip moment asymmetries between the intact and prosthetic sides were also reduced during both sit-to-stand (-0.18 ± 0.09 N/kg vs. -0.69 ± 0.67 N/kg) and stand-to-sit transitions (0.77 ± 0.20 N/kg vs. 2.03 ± 0.58 N/kg). Conclusions: The prototype and control strategy demonstrated promising improvements in sitting and standing performance compared to conventional passive prostheses, reducing the physical demand on the intact limb and upper body.