Medical Engineering & Physics最新文献

Sleep quality is fundamentally linked to health, and prolonged static postures during sleep can induce muscular fatigue, discomfort, and musculoskeletal risk. A critical gap exists in understanding the continuous relationship between muscle state dynamics and body-mattress interface pressure throughout sleep, which is essential for developing ergonomically adaptive sleep systems. Ten healthy adults maintained a supine posture on a standardized mattress equipped with a high-resolution pressure array. Surface electromyography was recorded from the trapezius, erector spinae, gluteus medius, and biceps femoris muscles. Muscle states were classified using the joint analysis of EMG spectrum and amplitude (JASA) method across three hours of sleep. Regional pressure parameters were calculated for the shoulder, waist & back, hips, and thigh. Muscle states exhibited distinct regional and temporal patterns: trapezius activity remained stable (recovery/force decrease), erector spinae progressively shifted toward fatigue, gluteus medius transitioned from fatigue toward recovery, and biceps femoris showed increasing fatigue. Interface pressure was highest at the shoulder and hips. Statistically significant negative correlations were identified between muscle fatigue and regional pressure, most strongly between gluteus medius fatigue and hips pressure distribution (Kendall'sτ=  -0.723,p< 0.01). Reduced interface pressure in partially suspended body regions appears to elicit compensatory muscle activation, leading to fatigue accumulation. The JASA method effectively discriminates muscle states during prolonged supine posture, providing a biomechanical basis for real-time, pressure-modulating interventions aimed at improving sleep recovery and comfort.

Accurate identification of the epidural space is critical for procedures such as labor analgesia, postoperative pain management, and epidural steroid injections. The current loss-of-resistance (LOR) technique depends on subtle tactile cues, which are highly subjective and prone to variability and complications. The objective was to develop and evaluate a prototype device (EpiduraFlow) that provides real-time quantitative feedback using pressure and flow metrics to enhance the accuracy and reliability of epidural space identification. A prototype system was designed incorporating a piezoelectric micropump, differential pressure sensors, and a microcontroller with LCD display. The device infused saline at a controlled rate through a Tuohy epidural needle, continuously recording flow and pressure. Testing was performed on a validated epidural simulation model at the UCSD Health SimCenter. Flow and pressure changes were analyzed during needle advancement across simulated tissue layers. Mean flow rate during advancement through simulated soft tissue and ligaments layers was 1.02 ± 0.84 µl s^-1, compared with 29.7 ± 5.3 µl s^-1upon entry into the simulated epidural space (p≪0.001). Pressure dropped correspondingly at the moment of entry, and changes were displayed in real time on the LCD. Calibration of sensors against a manometer demonstrated high linearity (R2>0.98). EpiduraFlow reliably identified transitions into the epidural space during simulated procedures. This proof-of-concept demonstrates the feasibility of objective, quantitative epidural localization and supports further development toward handheld, sterile-compatible designs and preclinical validation.

Fast and accurate detection of abnormalities, such as tumor nodules, in soft tissue is a critical step toward effective cancer diagnosis. Clinical examples include the use of tactile feedback during digital rectal examination for prostate cancer screening and intraoperative tumor localization.However, the absence of robust mechanical characterization and detection methods limits the clinical applicability of these techniques. In this study, we investigate instrumented indentation as a tool for detecting tumor-mimicking nodules embedded within porcine liver tissue models. Multi mechanical characterization, including hyperelasticity, viscoelasticity, and dynamic indentation, was performed to capture the mechanical response of the tissue at different points across its surface. A multivariate statistical outlier detection approach, based on Mahalanobis distance, was applied to assess the effectiveness of different mechanical metrics in identifying embedded nodules. The results demonstrate that this outlier detection framework reliably identifies stiff nodules within one to two standard deviations, offering a promising, clinically relevant method for soft tissue cancer detection.

Hydrocephalus is a condition that can result in increased intracranial pressure due to the excessive accumulation of cerebrospinal fluid (CSF) or blockage of the outflow of CSF at the level of the third or fourth ventricle. The basic treatment for hydrocephalus is CSF drainage by inserting a surgical tube (shunt) into the ventricle. Despite many new shunt design solutions, shunt failures are a regular occurrence, further complicating treatment. To address this problem, we demonstrate the effectiveness of a robust multi-objective design method to optimize the design of a canine ventricular shunt. In this paper, we investigate the causes of shunt failures and utilize this knowledge to obtain design solutions that minimize the risk of failures in shunts. Key shunt performance parameters that influence shunt failures, such as flow velocity, pressure difference and shear stress, are identified and their relationships with shunt geometry are established. A multi-objective robust design problem is formulated to identify shunt designs that are relatively insensitive to uncertainties while satisfying multiple, conflicting design goals. By using a multi-objective robust formulation, (1) the risk of shunt failures is minimized by designing against multiple criteria that govern failures in shunt, and (2) the reliability of the undertaken design decisions is improved by identifying and managing uncertainties. The method presented is demonstrated using a canine shunt design. However, it is generic enough to be applied to the design of human ventricular shunts and similar biomedical devices.

Osseointegrated transfemoral prostheses improve mobility but may induce stress shielding, periprosthetic bone loss, and long-term mechanical complications. Because amputees exhibit substantial variability in femoral geometry and cortical density, understanding how bone quality affects postoperative adaptation is essential. This study used three-dimensional finite element models coupled with a strain-energy-based remodeling algorithm to investigate how variations in femoral shaft diameter and initial apparent density influence stress transfer, density adaptation, and failure risk around the implant. Nine femur models were generated by combining three shaft diameters (24-28 mm) with three initial density levels. Remodeling was simulated over 60 months under physiological loading. Low-density femurs exhibited substantial proximal densification and pronounced distal bone loss, accompanied by elevated failure risk at the boneimplant interface. High-density femurs showed minimal remodeling and consistently lower stress and risk levels. Bone shaft diameter modulated, but did not override, the dominant effect of initial density. These findings highlight the importance of preoperative evaluation of cortical density and geometry when planning direct skeletal fixation in transfemoral amputees.

Standard x-ray radiography is routinely used to monitor total knee arthroplasty (TKA) postoperatively for complications such as loosening, malpositioning and insert wear. However, the radiolucency of the polyethylene insert makes quantitative wear assessment challenging. This study investigates the integration of radiopaque markers into standard ultra-high molecular weight polyethylene (UHMWPE) inserts to enhance their radiographic visibility and enable quantitative wear assessment from standard radiographs. Preliminary experiments established suitable process parameters for micro-milling cavities into UHMWPE. Final inserts were machined with varying microstructure configurations comprising grooves and holes. These microstructures were filled with a radiopaque composite of high-density polyethylene (HDPE) +20 wt.% barium sulphate (BaSO₄) composite via extrusion. HDPE was employed as a substitute for UHMWPE due to processability challenges resulting from the latter's high melt viscosity. The marker-integrated inserts were successively fitted on a phantom knee setup fitted together with TKA components and radiographed in the anteroposterior view. A weighted scoring model was created to identify optimal marker geometries based on edge visibility, dimensional measurability, homogeneity, and implant-induced obscuration of the marker projections in standard radiographs. Vertical groove markers i.e. those oriented in parallel to the central ray exhibited superior radiographic visibility and measurability compared to horizontal grooves. Hole markers exhibited a higher homogeneity and were easier to fill, but showed slightly reduced radiographic edge definition in comparison to the vertical grooves. Overall, the vertical grooves were identified as the most favourable marker geometry, followed by the holes, whereas horizontal markers performed the poorest. The findings of this study provide a proof of concept for incorporating radiopaque markers into TKA inserts, establishing a methodological framework for futurein vitrowear measurement based on dimensional marker change analysis. Further research into the development of a UHMWPE-compatible marker material is required before clinical relevance can be achieved.

The moment of inertia (MoI) is an important parameter in biomechanical modeling that can influence the accuracy of kinetic estimations. The most commonly used approach to estimate MoI relies on anthropometric tables derived from a limited number of subjects, which may not account for subject-specific variability. This study evaluated the previously proposed angular momentum technique for estimating subject-specific body MoI in two healthy adult males. We were unable to obtain realistic MoI estimates using the angular momentum method, with MoI values being up to 36 times larger than reference values. Initial investigations revealed two promising alternative methods that yielded more realistic MoI estimates.

The effect of viscous energy dissipation on hemolysis has been studied primarily in homogeneous shear flow scenarios, where a clear relationship between viscous energy dissipation and blood damage is observed. However, medical blood-contact devices often involve more complex flow states, such as turbulent flow, small and larger perturbations from average velocity, vortices, etc which may alter this relationship. This study investigates how varying flow regimes influence the relationship between viscous energy dissipation and hemolysis, using a shearing device that mimics conditions typical of cardiovascular prosthetics. The results suggest that while energy dissipation correlates with hemolysis in homogeneous shear flow, in more complex flow states, this relationship is influenced by additional factors such as flow disturbances and turbulence. Notably, at the same viscous energy dissipation rate, flow in disturbed states resulted in higher levels of hemolysis compared to homogeneous shear flow. These findings highlight the necessity of incorporating various flow conditions into future hemolysis prediction models to better understand and mitigate blood damage in medical device design.

Precise retinal vessel segmentation techniques are crucial for computer-aided clinical diagnosis. Recent advancements in deep learning have considerably enhanced segmentation accuracy (ACC); however, existing methods often struggle with thin and fuzzy boundaries due to cross-connection redundancy. Most mainstream models rely on complex encoders, resulting in high parameter counts and resource demands. Therefore, we propose LCNet, a lightweight U-shaped network with depth-separable convolution to minimize parameters and computational costs. It incorporates a synergistic coordinate attention module to enhance feature learning and captures multiscale features through the atrous spatial pyramid pooling module. Additionally, it introduces four side-output layers for extra supervision. Evaluating LCNet on four classical datasets (DRIVE, STARE, CHASEDB1, and IOSTAR), LCNet achieves global accuracies of 96.02%, 97.95%, 97.95%, and 97.77%. Our experiments that LCNet delivers enhanced performance with only 2.65 M parameters and 21.2 GFLOPs on DRIVE. We also demonstrate the effectiveness of LCNet on fundus images with lesions and optical coherence tomography angiography images, establishing it as a lightweight model with high efficiency and ACC for retinal vessel segmentation.

Accurate arrhythmia classification from short clinical electrocardiograms (ECGs) remains challenging, particularly in terms of performance and interpretability. Prior studies are often limited to single-lead, use uniform model fusion, or report point estimates without uncertainty. To address this, we propose a novel class-specific weighted ensemble for 12-lead, 10 s ECG segments that fuses multiple models with per-class weights and provides lead-resolved SHapley Additive exPlanations (SHAP) explanations over 360 features. Four rhythms (sinus bradycardia (SB), sinus tachycardia (ST), atrial flutter (AF), supraventricular tachycardia (SVT)) were evaluated on the Chapman-Shaoxing dataset using an estimation-based framework with Wilson and Newcombe 95% confidence intervals (CIs) and Cohen'sh. Compared with a Bagging baseline, the proposed ensemble yields higher SB recall on both validation and test (Δ +0.040 and +0.050; CIs entirely positive; smallh), higher ST precision (Δ +0.070 and +0.080; smallh), and comparable overall accuracy (ΔCIs spanning 0;h≈ 0.00-0.02). AF and SVT differences are small or statistically imprecise. SHAP highlights lead-specific contributors (e.g. P-wave area in V2), indicating a path toward minimal-lead configurations while preserving performance. Collectively, integrating per-class weighted fusion, lead-level interpretability, and CI-based estimation on short 12-lead ECGs yields class-specific gains (higher SB sensitivity and ST precision), and SHAP pinpoints lead V2-with P-wave area among the top contributors. The outcome of this study provides potential insights that may be used to classify arrhythmia more accurately for each class and develop minimal lead configurations, given the growing use of wearable and portable devices. However, further studies and data are required for clinical application and device development.