Objective: Conventional physical activity (PA) metrics derived from wearable sensors may not capture the cumulative, transitions from sedentary to active, and multidimensional patterns of PA, limiting the ability to predict physical function impairment (PFI) in older adults. This study aims to identify unique temporal patterns and develop novel digital biomarkers from wrist accelerometer data for predicting PFI and its subtypes using explainable artificial intelligence techniques. Materials and Methods: Wrist accelerometer streaming data from 747 participants in the National Health and Aging Trends Study (NHATS) were used to calculate 231 PA features through time-series analysis techniques—Tsfresh. Predictive models for PFI and its subtypes (walking, balance, and extremity strength) were developed using 6 machine learning (ML) algorithms with hyperparameter optimization. The SHapley Additive exPlanations method was employed to interpret the ML models and rank the importance of input features. Results: Temporal analysis revealed peak PA differences between PFI and healthy controls from 9:00 to 11:00 am. The best-performing model (Gradient boosting Tree) achieved an area under the curve score of 85.93%, accuracy of 81.52%, sensitivity of 77.03%, and specificity of 87.50% when combining wrist accelerometer streaming data (WAPAS) features with demographic data. Discussion: The novel digital biomarkers, including change quantiles, Fourier transform (FFT) coefficients, and Aggregated (AGG) Linear Trend, outperformed traditional PA metrics in predicting PFI. These findings highlight the importance of capturing the multidimensional nature of PA patterns for PFI. Conclusion: This study investigates the potential of wrist accelerometer digital biomarkers in predicting PFI and its subtypes in older adults. Integrated PFI monitoring systems with digital biomarkers would improve the current state of remote PFI surveillance.

Background: Minimally invasive surgery, in particular endoscopic surgery, has revolutionized the benefits for patients, but poses greater challenges for surgeons in terms of ergonomics. Integrating ergonomic assessments and interventions into the multi-stage endoscopic procedure contributes to the surgeon’s musculoskeletal health and the patient’s intraoperative safety and postoperative recovery.Objective: The purpose of this study was to overview the objective assessment techniques, tools and assessment settings involved in endoscopic procedures over the past decade and to identify the potential factors that induce differences in high workloads in endoscopic procedures and ultimately to design a framework for ergonomic assessment in endoscopic surgery.Methods: Literature searches were systematically conducted in the OVID, pubmed and web of science database before October 2022, and studies evaluating ergonomics during the process of endoscopic procedures or simulated procedures were both recognized.Results: Our systematic review of 56 studies underscores ergonomic variations in endoscopic surgery. While endoscopic procedures, predominantly laparoscopy, typically incur less physical load than open surgery, extended surgical durations notably elevate ergonomic risks. Surgeon characteristics, such as experience level and gender, significantly influence these risks, with less experienced and female surgeons facing greater challenges. Key assessment tools employed include electromyography for muscle fatigue and motion analysis for postural evaluation.Conclusion: This review aims to provide a comprehensive analysis and framework of objective ergonomic assessments in endoscopic surgery, and suggesting avenues for future research and intervention strategies. By improving the ergonomic conditions for surgeons, we can enhance their overall health, mitigate the risk of WMSDs, and ultimately improve patient outcomes.

Background: Physical activity (PA) is crucial for maintaining physical function in older adults, but relationships between multidimensional PA patterns and functional decline remain unclear. This study examined associations between accelerometer-measured PA patterns and physical function decline in older adults. Methods: We conducted a prospective cohort study with 1-year follow-up using data from 586 community-dwelling participants aged ≥65 years in the National Health and Aging Trends Study (2021-2022). Wrist-worn accelerometers measured 4 PA dimensions: cumulative (total activity counts), peak (maximum intensity), temporal (active and sedentary minutes), and fragmentation. Physical function decline was defined as any decrease in Short Physical Performance Battery score at follow-up. Multivariable logistic regression examined associations between PA dimensions and physical function changes. Results: Higher total activity counts (odds ratio [OR]: 0.71, 95% CI; 0.59-0.85), minutes spent active (OR: 0.75, 95% CI; 0.63-0.89), and maximum intensity (OR: 0.67, 95% CI; 0.55-0.83) were associated with lower odds of functional decline, while activity fragmentation showed the opposite relationship (OR: 1.23, 95% CI; 1.03-1.47). Dose-response analyses demonstrated continuous linear relationships. Compared to the lowest activity levels (10th percentile), participants at the 90th percentile showed substantially lower risk: total activity counts (OR: 0.50, 95% CI; 0.29-0.88), active minutes (OR: 0.58, 95% CI; 0.37-0.95), and maximum intensity (OR: 0.54, 95% CI; 0.32-0.75), while activity fragmentation showed progressive risk increase (OR: 1.37, 95% CI; 0.83-2.21). Domain-specific analyses showed consistent patterns. Conclusions: Multidimensional PA patterns have distinct relationships with functional decline in older adults. Findings support tailored PA recommendations and potential for targeted interventions.

Drug-drug interaction (DDI) refers to the inhibitory or enhancing effects between different drugs. Existing DDI prediction methods primarily use graph neural networks (GNNs) to directly represent drug molecular features. However, they often ignore the 3D structures of different atoms within drug molecules and the impact of noise in GNNs on DDI prediction. Consequently, the accuracy of GNN-based DDI prediction remains unsatisfactory. To address these limitations, this study proposes a DDI prediction method based on atomic 3D position encoding and an elastic message passing graph neural network (A3DPE-EMPGNN). Firstly, we construct an atomic feature network based on an attention mechanism and a message passing neural network. This network leverages 3D position encoding based on the molecular centroid to learn the features of different atoms and their associated chemical bonds, thereby constructing a graph-based molecular representation. Secondly, we design a molecular feature network that incorporates an attention mechanism, utilizing multi-head attention to capture interaction information between different drug molecules. Thirdly, we employ an adversarial attack detection and defense strategy, integrating supervised and contrastive loss learning to optimize the model and enhance its robustness while performing DDI prediction. Lastly, we evaluate the effectiveness of A3DPE-EMPGNN on two real-world datasets. Experimental results clearly demonstrate that our method achieves over 98% accuracy across ACC, AUC, AP, and F1-score, outperforming state-of-the-art GNN-based models.

Diagnosing physical function impairment (PFI) in older adults is challenging due to delayed diagnosis and cumbersome clinical assessments. We propose PFIMPA, a novel artificial intelligence method that enables seamless remote PFI assessment using physical activity (PA) data provided by wrist-worn accelerometers for the first time. To capture the complex cumulative and dynamic characteristics of PA, we designed a comprehensive feature representation method that integrates demographic information, clinical PA features, time-frequency domain features, and PA sequence conversion image features (PASCIs) based on Gramian Angular Field, Markov Transfer Field, and Recursive Plot transformations. We developed a lightweight convolutional neural network enhanced by Convolutional Block Attention Module to extract spatio-temporal features from PASCIs, further encapsulating the correlation, accumulation, and transition properties of PA. PFI detection is achieved through a three-stage feature-level fusion method combined with machine learning and hyperparameter optimization. The PASCIs module achieves an AUC of 83.72% in PFI prediction, while the multimodal approach combining clinical, time-frequency, and PASCIs features improves the AUC to 87.16%, surpassing state-of-the-art time series classification models. PFIMPA represents a significant advancement in PFI assessment for older adults, enabling seamless remote monitoring and timely intervention.

Early detection of Mild Cognitive Impairment (MCI) offers a critical window for intervention before progression to dementia. This study presents a novel approach to MCI screening through multimodal behavioral analysis during the Clock Drawing Test (CDT), implementing a comprehensive data collection system using standard tablet devices. Our approach analyzes three behavioral dimensions: drawing trajectories, facial expressions, and hand motion patterns. We developed an integrated framework combining Convolutional Neural Networks with multi-head attention mechanisms for drawing analysis, alongside specialized feature extraction methods for facial and motion data. To effectively integrate these behavioral markers, we proposed a Higher-order Canonical Correlation Analysis (HCCA) algorithm that models complex relationships between modalities. In our primary study with 69 MCI participants and 133 age-matched controls, we identified distinct behavioral patterns: temporal drawing anomalies, altered facial behaviors including increased gaze variability, and distinctive motion patterns reflecting changes in movement stability. Our multimodal approach achieved an AUC of 0.92 (accuracy: 92.68%, sensitivity: 90.91%), with HCCA further improving performance to an AUC of 0.97. External validation with 76 additional participants demonstrated good generalizability (AUC: 0.86). These findings reveal how cognitive impairment manifests across multiple behavioral channels offer clinicians precise, interpretable quantification of cognitive function for more accurate MCI screening.