Annals of Biomedical Engineering最新文献

The ability of articular cartilage to withstand significant mechanical stresses during activities, such as walking or running, relies on its distinctive structure. Integrating detailed tissue properties into subject-specific biomechanical models is challenging due to the complexity of analyzing these characteristics. This limitation compromises the accuracy of models in replicating cartilage function and impacts predictive capabilities. To address this, methods revealing cartilage function at the constituent-specific level are essential. In this study, we demonstrated that computational modeling derived individual constituent-specific biomechanical properties could be predicted by a novel nanoparticle contrast-enhanced computer tomography (CECT) method. We imaged articular cartilage samples collected from the equine stifle joint (n = 60) using contrast-enhanced micro-computed tomography (µCECT) to determine contrast agents’ intake within the samples, and compared those to cartilage functional properties, derived from a fibril-reinforced poroelastic finite element model. Two distinct imaging techniques were investigated: conventional energy-integrating µCECT employing a cationic tantalum oxide nanoparticle (Ta₂O₅-cNP) contrast agent and novel photon-counting µCECT utilizing a dual-contrast agent, comprising Ta₂O₅-cNP and neutral iodixanol. The results demonstrate the capacity to evaluate fibrillar and non-fibrillar functionality of cartilage, along with permeability-affected fluid flow in cartilage. This finding indicates the feasibility of incorporating these specific functional properties into biomechanical computational models, holding potential for personalized approaches to cartilage diagnostics and treatment.

Heart failure with preserved ejection fraction (HFpEF) constitutes approximately 50% of heart failure (HF) cases, and encompasses different phenotypes. Among these, most patients with HFpEF exhibit structural heart changes, often with smaller left ventricular cavities, which pose challenges for utilizing ventricular assist devices (VADs). A left atrial to aortic (LA-Ao) VAD configuration could address these challenges, potentially enhancing patient quality of life by lowering elevated mean left atrial pressure (MLAP). This study assessed the anatomical compatibility and left atrial unloading capacity using a simulated VAD-supported HFpEF patient. A HeartMate3-supported HFpEF patient in an LA-Ao configuration was simulated using a cardiovascular simulator. Hemodynamic parameters were recorded during rest and exercise at seven pump flow rates. Computed tomography scans of 14 HFpEF (NYHA II–III) and six heart failure with reduced ejection fraction patients were analysed for anatomical comparisons. HFpEF models were independently assessed for virtual anatomical fit with the HM3 in the LA-Ao configuration. Baseline MLAP was reduced from 15 to 11 mmHg with the addition of 1 L/min HM3 support in the rest condition. In an exercise simulation, 6 L/min of HM3 support was required to reduce the MLAP from 29 to 16 mmHg. The HM3 successfully accommodated six HFpEF patients without causing interference with other cardiac structures, whereas it caused impingement ranging from 4 to 14 mm in the remaining patients. This study demonstrated that the HM3 in an LA-Ao configuration may be suitable for unloading the left atrium and relieving pulmonary congestion in some HFpEF patients where size-related limitations can be addressed through pre-surgical anatomical fit analysis.

Catastrophic cervical spine injuries in rugby often occur during tackling. The underlying mechanisms leading to these injuries remain unclear, with neck hyperflexion and buckling both proposed as the causative factor in the injury prevention literature. The aim of this study was to evaluate the effect of pre-impact head–neck posture on intervertebral neck loads and motions during a head-first rugby tackle. Using a validated, subject-specific musculoskeletal model of a rugby player, and computer simulations driven by in vivo and in vitro data, we examined the dynamic response of the cervical spine under such impact conditions. The simulations demonstrated that the initial head–neck sagittal-plane posture affected intervertebral loads and kinematics, with an extended neck resulting in buckling and supraphysiologic intervertebral shear and flexion loads and motions, typical of bilateral facet dislocation injuries. In contrast, an initially flexed neck increased axial compression forces and flexion angles without exceeding intervertebral physiological limits. These findings provide objective evidence that can inform injury prevention strategies or rugby law changes to improve the safety of the game of rugby.

Currently, there are no methods or tools available in clinical practice for classifying future knee osteoarthritis (KOA). In this study, we aimed to fill this gap by classifying future KOA into three severity grades: KL01 (healthy), KL2 (moderate), and KL34 (severe) based on the Kellgren-Lawrance scale. Due to the complex nature of multiclass classification, we used a two-stage method, which separates the classification task into two binary classifications (KL01 vs. KL234 in the first stage and KL2 vs. KL34 in the second stage). Our machine learning (ML) model used two Balanced Random Forest algorithms and was trained with gender, age, height, weight, and quantitative knee morphology obtained from magnetic resonance imaging. Our training dataset comprised longitudinal 8-year follow-up data of 1213 knees from the Osteoarthritis Initiative. Through extensive experimentation with various feature combinations, we identified KL baseline and weight as the most essential features, while gender surprisingly proved to be one of the least influential feature. Our best classification model generated a weighted F1 score of 79.0% and a balanced accuracy of 65.9%. The area under the receiver operating characteristic curve was 83.0% for healthy (KL01) versus moderate (KL2) or severe (KL34) KOA patients and 86.6% for moderate (KL2) versus severe (KL34) KOA patients. We found a statistically significant difference in performance between our two-stage classification model and the traditional single-stage classification model. These findings demonstrate the encouraging results of our two-stage classification model for multiclass KOA severity classification, suggesting its potential application in clinical settings in future.

Sporting helmets contain force attenuating materials which reduce traumatic head injury risk and may influence sport-related concussion (SRC) sequelae. The purpose of this study was to examine the association of sport helmet status with SRC-clinical presentation and recovery trajectories in men’s collegiate athletes. Sport helmet status was based on the nature of sports being either helmeted/non-helmeted. 1070 SRCs in helmeted (HELM) sports (Men’s-Football, Ice Hockey, and Lacrosse), and 399 SRCs in non-helmeted (NOHELM) sports (Men’s-Basketball, Cheerleading, Cross Country/Track & Field, Diving, Gymnastics, Soccer, Swimming, Tennis, and Volleyball) were analyzed. Multivariable negative binomial regression models analyzed associations between sport helmet status and post-injury cognition, balance, and symptom severity, adjusting for covariate effects (SRC history, loss of consciousness, anterograde/retrograde amnesia, event type). Kaplan-Meier curves evaluated median days to: initiation of return to play (iRTP) protocol, and unrestricted RTP (URTP) by sport helmet status. Log-rank tests were used to evaluate differential iRTP/URTP between groups. Two independent multivariable Weibull accelerated failure time models were used to examine differential iRTP and URTP between groups, after adjusting for aforementioned covariates and symptom severity score. Overall, the median days to iRTP and URTP was 6.3 and 12.0, respectively, and was comparable across NOHELM- and HELM-SRCs. Post-injury symptom severity was lower (Score Ratio 0.90, 95%CI 0.82, 0.98), and cognitive test performance was higher (Score Ratio 1.03, 95%CI 1.02, 1.05) in NOHELM-compared to HELM-SRCs. Estimated time spent recovering to iRTP/URTP was comparable between sport helmet status groups. Findings suggest that the grouping of sports into helmeted and non-helmeted show slight differences in clinical presentation but not recovery.

Evaluating the bone–implant interface (BII) properties of osseointegrated transfemoral (TFA) implants is important for early failure detection and prescribing loads during rehabilitation. The objective of this work is to derive and validate a 1D finite element (FE) model of the Osseointegrated Prosthetic Limb (OPL) TFA system that can: (1) model its dynamic behaviour and (2) extract the BII properties. The model was validated by: (1) comparing the 1D FE formulation to the analytical and 3D FE solutions for a simplified cylinder, (2) comparing the vibration modes of the actual TFA geometry using 1D and 3D FE models, and (3) evaluating the BII properties for three extreme conditions (LOW, INTERMEDIATE, and HIGH) generated using 3D FE and experimental (where the implant was embedded, using different adhesives, in synthetic femurs) signals for additional validation. The modes predicted by the 1D FE model converged to the analytical and the 3D FE solutions for the cylinder. The 1D model also matched the 3D FE solution with a maximum frequency difference of 2.02% for the TFA geometry. Finally, the 1D model extracted the BII stiffness and the system’s damping properties for the three conditions generated using the 3D FE simulations and the experimental INTERMEDIATE and HIGH signals. The agreement between the 1D FE and the 3D FE solutions for the TFA geometry indicates that the 1D model captures the system’s dynamic behaviour. Distinguishing between the different BII conditions demonstrates the 1D model’s potential use for the non-invasive clinical evaluation of the TFA BII properties.

AI shaming refers to the practice of criticizing or looking down on individuals or organizations for using AI to generate content or perform tasks. AI shaming has emerged as a recent phenomenon in academia. This paper examines the characteristics, causes, and effects of AI shaming on academic writers and researchers. AI shaming often involves dismissing the validity or authenticity of AI-assisted work, suggesting that using AI is deceitful, lazy, or less valuable than human-only efforts. The paper identifies various profiles of individuals who engage in AI shaming, including traditionalists, technophobes, and elitists, and explores their motivations. The effects of AI shaming are multifaceted, ranging from inhibited technology adoption and stifled innovation to increased stress among researchers and missed opportunities for efficiency. These consequences may hinder academic progress and limit the potential benefits of AI in research and scholarship. Despite these challenges, the paper argues that academic writers and researchers should not be ashamed of using AI when done responsibly and ethically. By embracing AI as a tool to augment human capabilities and by being transparent about its use, academic writers and researchers can lead the way in demonstrating responsible AI integration.