Influenza-like illnesses (ILI), such as influenza and RSV, pose significant global health burdens, especially in febrile children under 6 years old. Differentiating these from bacterial infections based solely on clinical symptoms is challenging. While PCR tests are reliable, they are costly and time-consuming. An effective predictive tool would help doctors prioritize tests and guide parents on seeking emergency care for their febrile children. We collected data from 2,559 children who visited the hospital for ILI inspections. We developed XGBoost models, comparing nine different machine learning algorithms. Our AI-assisted diagnostic pipeline consists of two stages: Decision Support System for patients (DSS-P): An in-house model using sex, age, symptoms, and medical history to decide on hospital visits. Decision Support System for clinicians (DSS-C): An in-hospital model incorporating breath sound types and Chest X-ray results to determine the necessity of clinical tests. We tested various experimental settings, including the addition of RAT-tested samples and the combined consideration of influenza and RSV. The performance for influenza achieved an Area Under the Curve of 0.749 and 0.776, while RSV achieved 0.907 and 0.924 in DSS-P and DSS-C, respectively. We identified biomarkers, noting that most biomarkers had opposite effects for influenza and RSV. This study developed predictive models for influenza and RSV and explored their underlying mechanisms. An expectation tool to guide doctors in prioritizing tests or assisting parents in deciding on emergency care for their febrile child would be invaluable. Biomarker analysis performed can provide insight on clinical fields.

Prosthetic knee joints are at the forefront of medical innovation, serving as crucial tools in restoring mobility and enhancing the quality of life for individuals grappling with knee-related ailments like osteoarthritis and injuries. By faithfully replicating the intricate biomechanics of the natural knee, these devices empower recipients to regain lost physical capabilities and lead active, fulfilling lives. This paper presents a novel methodology employing advanced control techniques, including sliding mode control (SMC) and super-twisting sliding mode control (STSMC), to explore lower limb dynamics and effectively manage a two-part knee joint replacement. Through meticulous parameter optimization using a genetic algorithm (GA), guided by the integral time absolute error as the optimization objective, the controllers are finely tuned to maximize performance and responsiveness in real-world scenarios. The stability of the proposed controllers is thoroughly validated using mathematical analysis based on Lyapunov stability criteria. This ensures they perform robustly and can withstand disturbances. Comprehensive performance evaluations conducted via MATLAB/Simulink simulations offer valuable insights into the comparative efficacy of different control strategies under varying conditions, facilitating informed decision-making and refinement of prosthetic knee design. Real-time validation of the proposed methodology is achieved through a hardware-in-loop experimental setup featuring the advanced C2000 Delfino MCU F28379D Launchpad.

The automated identification of the first and second heart sounds (S1 and S2, respectively) in phonocardiogram (PCG) signals plays a pivotal role in the detection of heart valve diseases based on the known occurrence of heart murmurs between S1-S2 or S2-S1 in valve disorders. Traditional neural network-based methods cannot differentiate between heart sounds and background noise, leading to decreased accuracy in the identification of crucial cardiac events. Therefore, a deep learning-based segmentation on PCG signals that can distinguish S1 and S2 heart sounds with the Convolutional Fourier transform (CF) modules, which are two sequentially connected CF modules, was proposed in this study. Internal datasets, alongside the publicly available PhysioNet 2016 dataset, were used for the training and validation of the CF modules to ensure a robust comparison against existing state-of-the-art models, specifically the logistic regression-Hidden semi-Markov model (LR-HSMM). The efficacy of the CF modules was further evaluated using external datasets, including the PhysioNet 2022 and the Asan Medical Center (AMC) datasets. The CF modules exhibited superior robustness and accuracy in segmenting S1 and S2, achieving an average F1 score of 97.64% for S1 and S2 segmentation, which indicated better performance compared with that of the previous best model, LR-HSMM. The integration of the CF modules ensures the robust performance of PCG segmentation even amidst heart murmurs and background noise, significantly contributing to the advancement of cardiac diagnostics. All code is available at https://github.com/mi2rl/PCG_FTseg.

to the voiding function/deficit of patients with pelvic organ prolapse (POP) waits to be clarified, this study investigated if RSCP modifies voiding functions of POP patients by focusing on its impact on the outlet resistance-dependent voiding workload using pressure-volume analysis (PVA), a protocol thermodynamically assaying work expenditure by the bladder in voiding cycles.

Hair loss affects significant social and psychological well-being issues of the person. Thus, various drugs, ingredients, and technologies are being developed to overcome it. Minoxidil (MXD) is a representative hair loss treatment drug because it suppresses the production of dihydrotestosterone and induces vasodilation. However, since MXD has various side effects when used orally, it is more desirable to use it topically. In this work, we disclosed a new polymeric formulation (MXD@CP) based on citric acid (CA) dimethylsiloxane polymer (CP) for the effective transdermal delivery of MXD.

This study aims to determine whether the removal of the femoral neck system (FNS) after bony union affects the biomechanical stability of the femur. Considering the technical challenges and potential complications, including screw stripping reported in recent studies, the study explores whether its removal impacts stress distribution within the femur and increases the risk of complications, such as screw stripping. The femurs were grouped into Intact, Group U (healed fractures with FNS in place), and Group R (healed fractures with FNS removed). Subgroup analysis was performed using Pauwels' classification for fractures at 30, 50, and 70 degrees. Finite element analysis (FEA) was used to model and evaluate the biomechanical behavior. Material properties for the models were derived from the literature. No significant difference in biomechanical stability was observed between Group U and Group R across the fracture angles tested, indicating that removal of FNS does not compromise the structural integrity of the femur. However, the risk of screw stripping during removal requires consideration. Removing the femoral neck system (FNS) after fracture healing preserves the femur's biomechanical stability, regardless of fracture angle. However, increased stress at the distal locking screw suggests caution to avoid complications such as screw stripping.

The concept of visual attention in dynamic contents, such as videos, has been much less studied than its counterpart, i.e., visual salience. Yet, temporal visual attention is useful for many downstream tasks, such as video compression and summarisation, or monitoring users' engagement with visual information. Previous work has considered quantifying a temporal salience score from spatio-temporal user agreements from gaze data. Instead of gaze-based or content-based approaches, we explore to what extent only brain signals can reveal temporal visual attention. We propose methods for (1) computing a temporal salience score from salience maps of video frames; (2) quantifying the temporal salience score as a cognitive consistency score from the brain signals from multiple observers; and (3) assessing the correlation between both temporal salience scores, and computing its relevance. Two public EEG datasets (DEAP and MAHNOB) are used for experimental validation. Relevant correlations between temporal visual attention and EEG-based inter-subject consistency were found, as compared with a random baseline. In particular, effect sizes, measured with Cohen's , ranged from very small to large in one dataset, and from medium to very large in another dataset. Brain consistency among subjects watching videos unveils temporal visual attention cues. This has relevant practical implications for analysing attention for visual design in human-computer interaction, in the medical domain, and in brain-computer interfaces at large.

Electrocardiogram (ECG) is mainly utilized for diagnosing heart diseases. However, various noises can influence the diagnostic accuracy. This paper presents a novel algorithm for denoising ECG signals by employing the Controlled Energy Constraint-Variational Mode Decomposition (CEC-VMD). Firstly, the noisy ECG signal is decomposed using CEC-VMD to obtain a set of intrinsic mode functions (IMFs) and a residual r. A modulation factor is utilized to minimize the modal information contained in the decomposed residuals. Furthermore, this paper presents an update formula for the modal and central frequencies based on ADMM. Finally, all the IMFs are integrated to obtain the ECG signal after denoising. By varying the value of the modulation factor, not only is the spectral energy loss of each mode reduced, but the orthogonality between the modes is also improved to better concentrate the energy of each mode. The experiments on simulated signals and MIT-BIH signals show that the average SNR after CEC-VMD denoising is 22.5139, the RMSE is 0.1128, and the CC is 0.9882. In addition, the proposed algorithm effectively improves the classification accuracy values, which are 99.0% and 99.9% for the SVM and KNN classifiers, respectively. These values are improved compared with those of EMD, VMD, SWT, SVD-VMD, and VMD-SWT. The proposed CEC-VMD technique for denoising ECG signals removes noise and better preserves features.

Positron Emission Tomography (PET) systems with high spatial resolution and sensitivity suffer from reduced photon transmittance due to the high aspect ratio of scintillation crystals and the large refractive index (RI) difference at the crystal-photosensor boundary. This study aimed to enhance light extraction from the scintillation crystal to the photosensor by applying various nanopatterns on the crystal surface. Various nanopattern shapes, including line, circular, hexagonal, and tapered pyramid, were designed and simulated using Monte Carlo and finite-difference time-domain (FDTD) methods. The optimization focused on the nanostructure's diameter, width, height, period ratio, and RI. Light extraction gain was evaluated against a reference dataset with a 100 nm thick airgap between the crystal and photosensor. Nanopatterns significantly improved light transmission at the crystal-photosensor boundary, especially for scintillation photons entering at angles larger than the critical angle. Hole-type patterns showed superior performance with lower heights, larger period ratios, and RIs between 1.7 and 1.9. A maximum light extraction gain of 1.46 was achieved with a hole-type circular nanopattern with an RI of 1.7. Furthermore, our simulation results were experimentally validated through the preliminary development of a nanopattern applied to the GAGG crystal. Nanopattern on the crystal surface can effectively enhance light extraction to the photosensor. These findings were experimentally validated, confirming the potential of nanopatterns in improving PET system performance.

Deep learning has demonstrated remarkable performance across various domains. One of the techniques contributing to this success is data augmentation. The essence of data augmentation lies in synthesizing data while preserving accurate labels. In this research, we introduce a data augmentation technique optimized for electrocardiogram (ECG) data by focusing on the unique angles between precordial leads in 12-lead ECG, considering situations that may occur in a clinical environment. Subsequently, we utilize the proposed data augmentation technique to train a deep learning model for diagnosing atrial fibrillation or atrial flutter, generalized supraventricular tachycardia, first-degree atrioventricular block, left bundle branch block and myocardial infarction from ECG signals, and evaluate its performance to validate the effectiveness of the proposed method. Compared to other data augmentation methods, our approach demonstrated improved performance across various datasets and most tasks, thereby showcasing its potential to enhance diagnostic accuracy. Additionally, our method is simple to implement, offering a gain in total training time compared to other augmentation methods. This study holds the potential to positively advance further development in the fields of bio-signal processing and deep learning technology, addressing the issue of the lack of optimized data augmentation techniques applicable to ECG data in the future.

In chronic respiratory diseases, continuous self-monitoring of vital signs such as respiratory rate aids in the early detection of exacerbations. In recent years, the development of smart clothing, such as garments equipped with sensors to measure respiratory rate, has been a focus of research. However, the usability and adoption of smart clothing are often compromised owing to the discomfort caused by compression pressure during wear. This study developed smart clothing designed to measure respiratory rate using a low compression pressure. This was achieved by integrating a bending angle sensor, based on double-layer capacitance, into the rib cage and abdomen areas. The accuracy of the respiratory rate measurement was evaluated in 20 healthy male subjects without respiratory diseases. Breathing was measured while the subjects wore the smart clothing and performed breathing exercises in sitting, supine, and lateral postures, following a metronome set between 12 and 30 bpm. To assess accuracy, the respiratory rate measured by the smart clothing was compared with that measured by a spirometer. The recorded compression pressure was 0.77 ± 0.21 kPa, with no subjects reporting discomfort. Correlation coefficients for respiratory rate in the different postures ranged within 0.97-0.99. The mean difference between the smart clothing and spirometer measurements was less than 0.1 bpm. The low mean difference indicated that the proposed low compression pressure wearable respiration sensor, employing a bending angle sensor based on double-layer capacitance, could measure respiratory rate accurately without causing discomfort and within an acceptable error range.

Unethical attempts to misuse and overdose opioids have led to strict prescription limits, necessitating frequent hospital visits and prescriptions for long-term severe pain management. Therefore, this study aimed to develop a prototype wearable device that facilitates the extended delivery of opioid drugs while incorporating abuse-deterrent functionality, referred to as the abuse deterrent device (ADD).

Reproducible control of tissue delivery and retention of a therapeutic agent facilitates the maximization of the efficacy of pharmacotherapy. Despite the proposal of various chemical and physical methods for modulating drug delivery and retention, the choice of the physical modality for this purpose has been limited in clinical practice. Thus, this study proposes a novel strategy for modulating the retention of a model antigen in tissues by using non-tissue-damaging high-frequency ultrasoundAfter the injection of a fluorescently labeled model antigen, followed by brief treatment with high-frequency ultrasound (5-20 MHz), the clearance of the antigen was monitored using a real-time near-infrared (NIR) fluorescence imaging system in vivo. Further, ultrasound treatment increased tissue retention at the site of model antigen administration. The results suggested that high-frequency ultrasound could change the response to a macromolecule injected into the tissue, such as a vaccine, thereby modulating the immune response. The proposed ultrasound-based technology is translatable to clinical settings and benefits from well-established ultrasound technologies that have been employed in various medical applications for decades. Moreover, this approach can be broadly applied to enhance the therapeutic effects of current and future immune-mediated therapeutic systems.

This study aims to establish a methodology for classifying gait patterns in patients with hip osteoarthritis without the use of wearable sensors. Although patients with the same pathological condition may exhibit significantly different gait patterns, an accurate and efficient classification system is needed: one that reduces the effort and preparation time for both patients and clinicians, allowing gait analysis and classification without the need for cumbersome sensors like EMG or camera-based systems. The proposed methodology follows three key steps. First, ground reaction forces are measured in three directions-anterior-posterior, medial-lateral, and vertical-using a force plate during gait analysis. These force data are then evaluated through two approaches: trend similarity is assessed using the Pearson correlation coefficient, while scale similarity is measured with the Symmetric Mean Absolute Percentage Error (SMAPE), comparing results with healthy controls. Finally, Gaussian Mixture Models (GMM) are applied to cluster both healthy controls and patients, grouping the patients into distinct categories based on six quantified metrics derived from the correlation and SMAPE. Using the proposed methodology, 16 patients with hip osteoarthritis were successfully categorized into two distinct gait groups (Group 1 and Group 2). The gait patterns of these groups were further analyzed by comparing joint moments and angles in the lower limbs among healthy individuals and the classified patient groups. This study demonstrates that gait pattern classification can be reliably achieved using only force-plate data, offering a practical tool for personalized rehabilitation in hip osteoarthritis patients. By incorporating quantitative variables that capture both gait trends and scale, the methodology efficiently classifies patients with just 2-3 ms of natural walking. This minimizes the burden on patients while delivering a more accurate and realistic assessment. The proposed approach maintains a level of accuracy comparable to more complex methods, while being easier to implement and more accessible in clinical settings.

In this paper, we propose a novel graph-based data augmentation method that can generally be applied to medical waveform data with graph structures. In the process of recording medical waveform data, such as electrocardiogram (ECG), angular perturbations between the measurement leads exist due to imperfections in lead positions. The data samples with large angular perturbations often cause inaccuracy in algorithmic prediction tasks. We design a graph-based data augmentation technique that exploits the inherent graph structures within the medical waveform data to improve the F1 score by 1.44% over various tasks, models, and datasets. In addition, we show that Graph Augmentation improves model robustness by testing against adversarial attacks. Since Graph Augmentation is methodologically orthogonal to existing data augmentation techniques, they can be used in conjunction to further improve the final performance, resulting in a 2.47% gain of the F1 score. We believe that our Graph Augmentation method opens up new possibilities to explore in data augmentation.

Sarcopenia is a rapidly rising health concern in the fast-aging countries, but its demanding diagnostic process is a hurdle for making timely responses and devising active strategies. To address this, our study developed and evaluated a novel sarcopenia diagnosis system using Stimulated Muscle Contraction Signals (SMCS), aiming to facilitate rapid and accessible diagnosis in community settings. We recruited 199 adults from Wonju Severance Christian Hospital between July 2022 and October 2023. SMCS data were collected using surface electromyography sensors with the wearable device exoPill. Their skeletal muscle mass index, handgrip strength, and gait speed were also measured as the reference. Binary classification models were trained to classify each criterion for diagnosing sarcopenia based on the AWGS cutoffs. The binary classification models achieved high discriminative abilities with an AUC score near 0.9 in each criterion. When combining these criteria evaluations, the proposed sarcopenia diagnosis system performance achieved an accuracy of 89.4% in males and 92.4% in females, sensitivities of 81.3% and 87.5%, and specificities of 91.0% and 93.8%, respectively. This system significantly enhances sarcopenia diagnostics by providing a quick, reliable, and non-invasive method, suitable for broad community use. The promising result indicates that SMCS contains extensive information about the neuromuscular system, which could be crucial for understanding and managing muscle health more effectively. The potential of SMCS in remote patient care and personal health management is significant, opening new avenues for non-invasive health monitoring and proactive management of sarcopenia and potentially other neuromuscular disorders.

While tripping is the leading cause of injurious falls in older adults, the influence of ankle and hip proprioceptive information in balance recovery among older adults is still not clearly understood. The objective of this study was to assess the influence of ankle vs. hip proprioceptive information by altering muscle spindle performance using vibratory stimulation among older adults and healthy young control participants. Two groups of young ( =  20, age =  22.2 ± 3.1 years) and older adult (  =  33, age = 74.0 ± 3.8 years) participants were recruited and went through treadmill perturbation (sudden backward treadmill movement mimicking a trip), while they were equipped with vibratory devices (no vibration, and 40 and 80 Hz) on either ankle or hip muscles. Kinematics of the recovery were measures using motion sensors on lower extremities and the trunk. Results showed that vibratory stimulation on ankle significantly influenced balance recovery response (i.e., increased reaction time by 18% and increased recovery step length by 21%) among healthy young control, while it showed no effect when placed on hip muscles. On the other hand, while vibratory stimulation on ankle showed no effect on balance recovery among older adults, it significantly influenced balance recovery when applied to the hip muscles (i.e., increased reaction time by 12% and increased recovery step length by 10%). Current findings suggest that the role of ankle vs. hip proprioceptive information in balance recovery may change by aging. Findings may potentially be used for targeting the appropriate location for balance interventions and reducing the fall risk in older adults.

To investigate the biomechanical effects of medial collateral ligament (MCL) stiffness adjustments on knee kinematics-medial femoral rollback, femoral rotation, and joint contact forces-in mechanically aligned posterior-substituting (PS) total knee arthroplasty (TKA). A musculoskeletal model simulating squatting was developed using the AnyBody modeling system. A PS-TKA prosthesis was implanted, and MCL stiffness was modified in 20% increments. The effects on femoral rollback, femoral rotation, and joint forces were evaluated. Medial femoral rollback was not significantly affected by changes in MCL stiffness. However, when MCL stiffness exceeded 20% above normal, the pattern and magnitude of lateral femoral rollback were altered compared to other conditions. Increased MCL stiffness also altered internal-external femoral rotation and raised joint contact forces in the medial compartment. Muscle activity was largely unaffected by changes in MCL stiffness, although hamstring activity increased slightly during early flexion (0°-5°) when MCL stiffness exceeded 20%. Excessive MCL stiffness (over 20% above normal) affects lateral femoral rollback and increases joint contact forces, potentially elevating the risk of prosthetic wear. Maintaining MCL stiffness within physiological limits is critical for optimizing outcomes in varus knee TKA.

NAFLD is gaining recognition as a complex, multifactorial condition with suspected associations with endocrine disorders. This investigation employed MR analysis to explore the potential causality linking NAFLD to a spectrum of endocrine diseases, encompassing T1D, T2D, obesity, graves' disease, and acromegaly.

Osteoarthritis (OA) is a musculoskeletal disorder that affects weight-bearing joints like the hip, knee, spine, feet, and fingers. It is a chronic disorder that causes joint stiffness and leads to functional impairment. Knee osteoarthritis (KOA) is a degenerative knee joint disease that is a significant disability for over 60 years old, with the most prevalent symptom of knee pain. Radiography is the gold standard for the evaluation of KOA. These radiographs are evaluated using different classification systems. Kellgren and Lawrence's (KL) classification system is used to classify X-rays into five classes (Normal = 0 to Severe = 4) based on osteoarthritis severity levels. In recent years, with the advent of artificial intelligence, machine learning, and deep learning, more emphasis has been given to automated medical diagnostic systems or decision support systems. Computer-aided diagnosis is needed for the improvement of health-related information systems. This survey aims to review the latest advances in automated radiographic classification and detection of KOA using the KL system. A total of 85 articles are reviewed as original research or survey articles. This survey will benefit researchers, practitioners, and medical experts interested in X-rays-based KOA diagnosis and prediction.

Robotic systems rely on spatio-temporal information to solve control tasks. With advancements in deep neural networks, reinforcement learning has significantly enhanced the performance of control tasks by leveraging deep learning techniques. However, as deep neural networks grow in complexity, they consume more energy and introduce greater latency. This complexity hampers their application in robotic systems that require real-time data processing. To address this issue, spiking neural networks, which emulate the biological brain by transmitting spatio-temporal information through spikes, have been developed alongside neuromorphic hardware that supports their operation. This paper reviews brain-inspired learning rules and examines the application of spiking neural networks in control tasks. We begin by exploring the features and implementations of biologically plausible spike-timing-dependent plasticity. Subsequently, we investigate the integration of a global third factor with spike-timing-dependent plasticity and its utilization and enhancements in both theoretical and applied research. We also discuss a method for locally applying a third factor that sophisticatedly modifies each synaptic weight through weight-based backpropagation. Finally, we review studies utilizing these learning rules to solve control tasks using spiking neural networks.