Search Results (4,132)

Conventional academic warning systems in higher education often rely on end-of-semester grades, which severely limits opportunities for timely intervention. To address this, our interdisciplinary study developed and validated a comprehensive socio-technical framework that integrates social-cognitive theory with learning analytics. The framework combines educational data mining with culturally responsive, personalized interventions tailored to a non-Western context. A two-phase mixed-methods design was employed: first, predictive models were built using Learning Management System (LMS) data from 2,856 students across 64 courses. Second, a quasi-experimental trial (n = 48) was conducted to evaluate intervention efficacy. Historical academic performance, attendance, and assignment submission patterns were the strongest predictors, achieving a Balanced Area Under the Curve (AUC) of 0.85. The intervention, specifically adapted to Confucian educational values, yielded remarkable results: 73% of at-risk students achieved passing grades, with a large effect size for academic improvement (Cohen’s d = 0.91). These findings empirically validate a complete prediction–intervention–evaluation cycle, demonstrating how algorithmic predictions can be effectively integrated with culturally informed human support networks. This study advances socio-technical systems theory in education by bridging computer science, psychology, and educational research. It offers an actionable model for designing ethical and effective early warning systems that balance technological innovation with human-centered pedagogical values. Full article

Machine learning (ML) algorithms have promoted the development of predictive modeling of mineral prospectivity, enabling data-driven decision-making processes by integrating multi-source geological information, leading to efficient and accurate prediction of mineral exploration targets. However, it is challenging to conduct ML-based mineral prospectivity mapping (MPM) in under-explored areas where scarce data are available. In this study, the Narigongma district of the Qiangtang block in the Himalayan–Tibetan orogen was chosen as a case study. Five typical alterations related to porphyry mineralization in the study area, namely pyritization, sericitization, silicification, chloritization and propylitization, were extracted by remote sensing interpretation to enrich the data source for MPM. The extracted alteration evidences, combined with geological, geophysical and geochemical multi-source information, were employed to train the ML models. Four machine learning models, including artificial neural network (ANN), random forest (RF), support vector machine and logistic regression, were employed to map the Cu-polymetallic prospectivity in the study area. The predictive performances of the models were evaluated through confusion matrix-based indices and success-rate curves. The results show that the classification accuracy of the four models all exceed 85%, among which the ANN model achieves the highest accuracy of 96.43% and a leading Kappa value of 92.86%. In terms of predictive efficiency, the RF model outperforms the other models, which captures 75% of the mineralization sites within only 3.5% of the predicted area. A total of eight exploration targets were delineated upon a comprehensive assessment of all ML models, and these targets were further ranked based on the verification of high-resolution geochemical anomalies and evaluation of the transportation condition. The interpretability analyses emphasize the key roles of spatial proxies of porphyry intrusions and geochemical exploration in model prediction as well as significant influences everted by pyritization and chloritization, which accords well with the established knowledge about porphyry mineral systems in the study area. The findings of this study provide a robust ML-based framework for the exploration targeting in greenfield areas with good outcrops but low exploration extent, where fusion of a remote sensing technique and multi-source geo-information serve as an effective exploration strategy. Full article

Differentiating genuine hemiplegic gait (HG) in stroke survivors from hemiplegic-like gait voluntarily imitated by healthy adults (MHG) is essential for reliable assessment and intervention planning. Treadmill-based gait data were obtained from 79 participants—39 stroke patients (HG) and 40 healthy adults—instructed to mimic HG (MHG). Forty-eight spatiotemporal and force-related variables were extracted. Random Forest, support vector machine (SVM), and logistic regression classifiers were trained with (i) the full feature set and (ii) the 10 most important features selected via Random Forest Gini importance. Performance was assessed with 5-fold stratified cross-validation and an 80/20 hold-out test, using accuracy, F1-score, and the area under the receiver operating characteristic curve (AUC). All models achieved high discrimination (AUC > 0.93). The SVM attained perfect discrimination (AUC = 1.000, test set) with the full feature set and maintained excellent accuracy (AUC = 0.983) with only the top 10 features. Temporal asymmetries, delayed vertical ground reaction force peaks, and mediolateral spatial instability ranked highest in importance. Reduced-feature models showed negligible performance loss, highlighting their parsimony and interpretability. Supervised machine learning algorithms can accurately distinguish true hemiplegic gait from mimicked patterns using a compact subset of gait features. The findings support data-driven, time-efficient gait assessments for clinical neurorehabilitation and for validating experimental protocols that rely on gait imitation. Full article

A critical factor contributing to the burden of childhood asthma is the lack of effective self-management in homecare settings. Artificial intelligence (AI) and lung sound monitoring could help address this gap. Yet, existing AI-driven auscultation tools focus on wheeze detection and often rely on subjective human labels. To improve the early detection of asthma worsening in children in homecare setting, we trained and evaluated a Deep Learning model based on spirometry-labelled lung sounds recordings to detect asthma exacerbation. A single-center prospective observational study was conducted between November 2020 and September 2022 at a tertiary pediatric pulmonology department. Electronic stethoscopes were used to record lung sounds before and after bronchodilator administration in outpatients. In the same session, children also underwent spirometry, which served as the reference standard for labelling the lung sound data. Model performance was assessed on an internal validation set using receiver operating characteristic (ROC) curves. A total of 16.8 h of lung sound recordings from 151 asthmatic pediatric outpatients were collected. The model showed promising discrimination performance, achieving an AUROC of 0.763 in the training set, but performance in the validation set was limited (AUROC = 0.398). This negative result demonstrates that acoustic features alone may not provide sufficient diagnostic information for the early detection of asthma attacks, especially in mostly asymptomatic outpatients typical of homecare settings. It also underlines the challenges introduced by differences in how digital stethoscopes process sounds and highlights the need to define the severity threshold at which acoustic monitoring becomes informative, and clinically relevant for home management. Full article

Cardiovascular diseases (CVDs) remain the leading cause of global mortality, with ischemic heart disease (IHD) being the most prevalent and deadly subtype. The growing burden of IHD underscores the urgent need for effective early detection methods that are scalable and non-invasive. Heart Rate Variability (HRV), a non-invasive physiological marker influenced by the autonomic nervous system (ANS), has shown clinical relevance in predicting adverse cardiac events. This study presents a photoplethysmography (PPG)-based Zhurek IoT device, a custom-developed Internet of Things (IoT) device for non-invasive HRV monitoring. The platform’s effectiveness was evaluated using HRV metrics from electrocardiography (ECG) and PPG signals, with machine learning (ML) models applied to the task of early IHD risk detection. ML classifiers were trained on HRV features, and the Random Forest (RF) model achieved the highest classification accuracy of 90.82%, precision of 92.11%, and recall of 91.00% when tested on real data. The model demonstrated excellent discriminative ability with an area under the ROC curve (AUC) of 0.98, reaching a sensitivity of 88% and specificity of 100% at its optimal threshold. The preliminary results suggest that data collected with the “Zhurek” IoT devices are promising for the further development of ML models for IHD risk detection. This study aimed to address the limitations of previous work, such as small datasets and a lack of validation, by utilizing real and synthetically augmented data (conditional tabular GAN (CTGAN)), as well as multi-sensor input (ECG and PPG). The findings of this pilot study can serve as a starting point for developing scalable, remote, and cost-effective screening systems. The further integration of wearable devices and intelligent algorithms is a promising direction for improving routine monitoring and advancing preventative cardiology. Full article

In-cylinder pressure provides critical insights for analyzing and optimizing combustion in internal combustion engines, yet its acquisition across the full operating space requires extensive testing, while physics-based models are computationally demanding. Machine learning (ML) offers an alternative, but its application to direct reconstruction of full pressure traces remains limited. This study evaluates three strategies for reconstructing cylinder pressure under unmeasured operating conditions, establishing a machine learning-assisted framework that generates the complete pressure–crank angle (P–CA) trace. The framework treats crank angle and operating conditions as inputs and predicts either pressure directly or apparent heat release rate (HRR) as an intermediate variable, which is then integrated to reconstruct pressure. In all approaches, discrete pointwise predictions are combined to form the full P–CA curve. Direct pressure prediction achieves high accuracy for overall traces but underestimates HRR-related combustion features. Training on HRR improves combustion representation but introduces baseline shifts in reconstructed pressure. A hybrid approach, combining non-combustion pressure prediction with combustion-phase HRR-based reconstruction delivers the most robust and physically consistent results. These findings demonstrate that ML can efficiently reconstruct in-cylinder pressure at unmeasured conditions, reducing experimental requirements while supporting combustion diagnostics, calibration, and digital twin applications. Full article

Background: Early detection of dysglycemia in young adults is important but underexplored. This study aimed to (1) predict long-term changes in fasting plasma glucose (δ-FPG) and (2) classify future prediabetes using complementary machine learning (ML) approaches. Methods: We analyzed 6247 Taiwanese men aged 18–35 years (mean follow-up 5.9 years). For δ-FPG (continuous outcome), random forest, stochastic gradient boosting (SGB), eXtreme gradient boosting (XGBoost), and elastic net were compared with multiple linear regression using Symmetric mean absolute percentage error (SMAPE), Root mean squared error (RMSE), Relative absolute error(RAE), and Root relative squared error (RRSE) Sensitivity analyses excluded baseline FPG (FPG_base). Shapley additive explanations(SHAP) values provided interpretability, and stability was assessed across 10 repeated train–test cycles with confidence intervals. For prediabetes (binary outcome), an XGBoost classifier was trained on top predictors, with class imbalance corrected by SMOTE-Tomek. Calibration and decision-curve analysis (DCA) were also performed. Results: ML models consistently outperformed regression on all error metrics. FPG_base was the dominant predictor in full models (100% importance). Without FPG_base, key predictors included body fat, white blood cell count, age, thyroid-stimulating hormone, triglycerides, and low-density lipoprotein cholesterol. The prediabetes classifier achieved accuracy 0.788, precision 0.791, sensitivity 0.995, ROC-AUC 0.667, and PR-AUC 0.873. At a high-sensitivity threshold (0.2892), sensitivity reached 99.53% (specificity 47.46%); at a balanced threshold (0.5683), sensitivity was 88.69% and specificity was 90.61%. Calibration was acceptable (Brier 0.1754), and DCA indicated clinical utility. Conclusions: FPG_base is the strongest predictor of glycemic change, but adiposity, inflammation, thyroid status, and lipids remain informative. A dual interpretable ML framework offers clinically actionable tools for screening and risk stratification in young men. Full article

Background and Objectives: To date, several machine learning (ML) prognostic prediction models have been investigated for patients with acute myocardial infarction (AMI). However, few studies have compared the prognostic performance of ML techniques in AMI patients who underwent percutaneous coronary intervention (PCI). We sought to compare the prognostic performance among various machine learning techniques to determine which one showed the best prediction ability. Materials and Methods: Using data from the large, multicenter COREA-AMI registry, this study analyzed 10,172 patients to predict major adverse cardiac events (MACEs) at 1 and 5 years. MACE was defined as a composite of cardiac death, myocardial infarction, or cerebrovascular accident. Results: Compared with the four other ML techniques and traditional logistic regression, the random forest (RF) model consistently demonstrated the highest predictive performance. At 5 years, the RF model achieved a superior area under the curve (AUC) of 0.822, an accuracy of 0.804, and an F1 score of 0.870. To ensure clinical interpretability, a SHapley Additive exPlanations analysis was performed on the RF model. It identified key independent predictors for MACEs. The top nonmodifiable predictors included age, renal function, and left ventricular ejection fraction, whereas modifiable risk factors included dual antiplatelet therapy, statin therapy, angiotensin-converting enzyme inhibitor/angiotensin receptor blocker therapy, and adherence to these optimal medical therapy. Conclusions: In this real-world patient cohort, the RF model provided modest improvements in long-term risk stratification, and our findings highlight the continuing importance of guideline-directed medical therapy in determining patient prognosis. Full article

With accelerating urbanization and global climate warming, Urban Heat Islands (UHIs) pose serious threats to urban development. Existing UHI research mainly focuses on inland regions, lacking systematic understanding of coastal city heat island mechanisms. We selected eight Chinese coastal cities with different backgrounds, quantitatively assessed urban heat island intensity based on summer 2023 Landsat 8 remote sensing data, established block-LCZ spatial analysis units, and employed a combination of machine learning models and causal inference methods to systematically analyze the regional differentiation characteristics of Urban Heat Island Intensity (UHII) and the influence mechanisms of multi-dimensional driving factors within land–sea interaction contexts. The results revealed the following: (1) UHII in the study area presents obvious spatial differentiation, with the highest value occurring in Hong Kong (2.63 °C). Northern cities generally had higher values than southern ones. (2) Different Local Climate Zone (LCZ) types show significant differences in thermal contributions, with LCZ2 (compact midrise) blocks presenting the highest UHII values in most cities, while LCZ G (water) and LCZ A (dense trees) blocks exhibit stable cooling effects. Nighttime light (NTL) and distance to sea (DS) are dominant factors affecting UHII, with NTL marginal effect curves generally presenting hump-shaped characteristics, while DS shows different response patterns across cities. (3) Causal inference reveals true causal driving mechanisms beyond correlations, finding that causal effects of key factors exhibit significant spatial heterogeneity. The research findings provide a new cognitive framework for understanding the formation mechanisms of thermal environments in Chinese coastal cities and offer a quantitative basis for formulating regionalized UHI mitigation strategies. Full article

Arthritis and arthrosis are prevalent joint diseases that cause pain and disability, and their early diagnosis is crucial for preventing irreversible damage. Conventional diagnostic methods such as X-ray, ultrasound, and MRI have limitations in early detection, prompting interest in alternative techniques. This work presents the design and clinical evaluation of a prototype device for non-invasive early diagnosis of arthritis (inflammatory joint disease) and arthrosis (osteoarthritis) using infrared thermography and deep neural networks. The portable prototype integrates a Raspberry Pi 4 microcomputer, an infrared thermal camera, and a touchscreen interface, all housed in a 3D-printed PLA enclosure. A custom Flask-based application enables two operational modes: (1) thermal image acquisition for training data collection, and (2) automated diagnosis using a pre-trained ResNet50 deep learning model. A clinical study was conducted at a university clinic in a temperature-controlled environment with 100 subjects (70% with arthritic conditions and 30% healthy). Thermal images of both hands (four images per hand) were captured for each participant, and all patients provided informed consent. The ResNet50 model was trained to classify three classes (healthy, arthritis, and arthrosis) from these images. Results show that the system can effectively distinguish healthy individuals from those with joint pathologies, achieving an overall test accuracy of approximately 64%. The model identified healthy hands with high confidence (100% sensitivity for the healthy class), but it struggled to differentiate between arthritis and arthrosis, often misclassifying one as the other. The prototype’s multiclass ROC (Receiver Operating Characteristic) analysis further showed excellent discrimination between healthy vs. diseased groups (AUC, Area Under the Curve ~1.00), but lower performance between arthrosis and arthritis classes (AUC ~0.60–0.68). Despite these challenges, the device demonstrates the feasibility of AI-assisted thermographic screening: it is completely non-invasive, radiation-free, and low-cost, providing results in real-time. In the discussion, we compare this thermography-based approach with conventional diagnostic modalities and highlight its advantages, such as early detection of physiological changes, portability, and patient comfort. While not intended to replace established methods, this technology can serve as an early warning and triage tool in clinical settings. In conclusion, the proposed prototype represents an innovative application of infrared thermography and deep learning for joint disease screening. With further improvements in classification accuracy and broader validation, such systems could significantly augment current clinical practice by enabling rapid and non-invasive early diagnosis of arthritis and arthrosis. Full article

Early and accurate diagnosis of dementia is essential to improving patient outcomes and reducing societal burden. The Mini-Mental State Examination (MMSE) is widely used to assess cognitive function, yet traditional statistical and machine learning approaches often face limitations in capturing nonlinear interactions and subtle decline patterns. This study developed a novel deep learning-based dementia prediction model using MMSE data collected from domestic clinical settings and compared its performance with traditional machine learning models. A notable strength of this work lies in its use of item-level MMSE features combined with explainable AI (SHAP analysis), enabling both high predictive accuracy and clinical interpretability—an advancement over prior approaches that primarily relied on total scores or linear modeling. Data from 164 participants, classified into cognitively normal, mild cognitive impairment (MCI), and dementia groups, were analyzed. Individual MMSE items and total scores were used as input features, and the dataset was divided into training and validation sets (8:2 split). A fully connected neural network with regularization techniques was constructed and evaluated alongside Random Forest and support vector machine (SVM) classifiers. Model performance was assessed using accuracy, F1-score, confusion matrices, and receiver operating characteristic (ROC) curves. The deep learning model achieved the highest performance (accuracy 0.90, F1-score 0.90), surpassing Random Forest (0.86) and SVM (0.82). SHAP analysis identified Q11 (immediate memory), Q12 (calculation), and Q17 (drawing shapes) as the most influential variables, aligning with clinical diagnostic practices. These findings suggest that deep learning not only enhances predictive accuracy but also offers interpretable insights aligned with clinical reasoning, underscoring its potential utility as a reliable tool for early dementia diagnosis. However, the study is limited by the use of data from a single clinical site with a relatively small sample size, which may restrict generalizability. Future research should validate the model using larger, multi-institutional, and multimodal datasets to strengthen clinical applicability and robustness. Full article

Background: Uniportal video-thoracoscopic lobectomy has improved postoperative outcomes in lung cancer patients. Thus, thoracic surgeons are increasingly required to learn this new approach. Methods: We evaluate the path of a single surgeon switching from triportal video-thoracoscopic lobectomy to the uniportal, using the cumulative sum (CUSUM) analysis, in a single center to assess the learning curve, enrolling 107 uniportal video-thoracoscopic lobectomies consecutively performed. CUSUM analysis detected how many uniportal video-thoracoscopies occur to obtain changes in mean operation time, among all procedures consecutively performed. CUSUM analysis identified the cut-off at the 67th procedure; this value was used to divide all patients into two groups: group A (first 67 patients, early phase) and group B (40 patients, experienced phase). Then, we analyze the perioperative outcomes between the two groups. Results: Gender characteristics of the two groups were statistically similar. Median operative time decreased significantly after the early phase [188 min (IQR: 151–236) vs. 170.5 (IQR: 134–202) (p-value = 0.02)], respectively. Similarly, during the second phase, the conversions rate decreased: [10 (15%) (group A) vs. 1 (2%) (group B) (p-value = 0.04)], as did the postoperative complications [28 cases (42%) vs. 9 cases (22%) (p-value = 0.04)] and the length of stay [6 days (IQR 5–9.5) vs. 5 days (IQR 4–8) (p-value = 0.04)], giving evidence of skills acquired in the second phase. Conclusions: CUSUM analysis identified 67 uniportal lobectomies, after which operative time, conversion rate, and perioperative complications significantly decreased; the moving average analysis further supports a progressive reduction in operative time. Despite prior multiportal video-thoracoscopic experience, switching to uniportal video-thoracoscopy requires a distinct learning process. Full article

Emerging applications like augmented reality (AR) demand efficient wireless transmission of high-resolution three-dimensional (3D) images, yet conventional systems struggle with the high data volume and vulnerability to noise. This paper proposes a novel feature-driven framework that integrates semantic source coding with deep learning-based Joint Source–Channel Coding (JSCC) for robust and efficient transmission. Instead of processing dense meshes, the method first extracts a compact set of geometric features—specifically, the ridge and valley curves that define the object’s fundamental structure. This feature representation which is extracted by the anatomical curves is then processed by an end-to-end trained JSCC encoder, mapping the semantic information directly to channel symbols. This synergistic approach drastically reduces bandwidth requirements while leveraging the inherent resilience of JSCC for graceful degradation in noisy channels. The framework demonstrates superior reconstruction fidelity and robustness compared to traditional schemes, especially in low signal-to-noise ratio (SNR) regimes, enabling practical and efficient 3D semantic communications. Full article

Background: The high-precision analysis of multimode fibers (MMFs) is a critical task in numerous applications, including remote sensing, medical imaging, and environmental monitoring. In this study, we propose a novel deep interpretable network approach to reconstruct spectral images captured using CCD sensors. Methods: Our model leverages a Tiny-YOLO-inspired convolutional neural network architecture, specifically designed for spectral wavelength prediction tasks. A total of 1880 CCD interference images were acquired across a broad near-infrared range from 1527.7 to 1565.3 nm. To ensure precise predictions, we introduce a dynamic factor α and design a dynamic adaptive loss function based on Huber loss and Log-Cosh loss. Results: Experimental evaluation with five-fold cross-validation demonstrates the robustness of the proposed method, achieving an average validation MSE of 0.0149, an R² score of 0.9994, and a normalized error (μ) of 0.0005 in single MMF wavelength prediction, confirming its strong generalization capability across unseen data. The reconstructed outputs are further visualized as smooth spectral curves, providing interpretable insights into the model’s decision-making process. Conclusions: This study highlights the potential of deep learning-based interpretable networks in reconstructing high-fidelity spectral images from CCD sensors, paving the way for advancements in spectral imaging technology. Full article

Accurate recognition of estrus behavior in sows is of great importance for achieving scientific breeding management, improving reproductive efficiency, and reducing labor costs in modern pig farms. However, due to the evident spatiotemporal continuity, stage-specific changes, and ambiguous category boundaries of estrus behaviors, traditional methods based on static images or manual observation suffer from low efficiency and high misjudgment rates in practical applications. To address these issues, this study follows a video-based behavior recognition approach and designs three deep learning model structures: (Convolutional Neural Network combined with Long Short-Term Memory) CNN + LSTM, (Three-Dimensional Convolutional Neural Network) 3D-CNN, and (Convolutional Neural Network combined with Temporal Convolutional Network) CNN + TCN, aiming to achieve high-precision recognition and classification of four key behaviors (SOB, SOC, SOS, SOW) during the estrus process in sows. In terms of data processing, a sliding window strategy was adopted to slice the annotated video sequences, constructing image sequence samples with uniform length. The training, validation, and test sets were divided in a 6:2:2 ratio, ensuring balanced distribution of behavior categories. During model training and evaluation, a systematic comparative analysis was conducted from multiple aspects, including loss function variation (Loss), accuracy, precision, recall, F1-score, confusion matrix, and ROC-AUC curves. Experimental results show that the CNN + TCN model performed best overall, with validation accuracy exceeding 0.98, F1-score approaching 1.0, and an average AUC value of 0.9988, demonstrating excellent recognition accuracy and generalization ability. The 3D-CNN model performed well in recognizing short-term dynamic behaviors (such as SOC), achieving a validation F1-score of 0.91 and an AUC of 0.770, making it suitable for high-frequency, short-duration behavior recognition. The CNN + LSTM model exhibited good robustness in handling long-duration static behaviors (such as SOB and SOS), with a validation accuracy of 0.99 and an AUC of 0.9965. In addition, this study further developed an intelligent recognition system with front-end visualization, result feedback, and user interaction functions, enabling local deployment and real-time application of the model in farming environments, thus providing practical technical support for the digitalization and intelligentization of reproductive management in large-scale pig farms. Full article