Bioengineering最新文献

Pain and variable uptake remain practical barriers to needle-based delivery. Device-level vibration has emerged as a simple strategy for improving tolerability and dispersion, but its fluid-mechanical basis remains incomplete. Using a lattice Boltzmann model with a porous-media skin surrogate, we applied time-periodic inlet pressures at 0%, 16.6% (ΔP1), and 35.1% (ΔP2) amplitudes to Newtonian, model shear-thinning, and clinically measured protein formulations. We quantified the wall shear stress, wetted area, dispersion length, and pressure cost over one cycle. Vibration increased the normalized wetted area by 10.6% for Newtonian flow and by 15.9% and 21.3% for the non-Newtonian cases at ΔP1 and ΔP2, respectively, while advancing the penetration front and lateral dispersion. The one-cycle pressure cost per wetted area decreased by 3.9% for Newtonian flow and by 5.96% and 7.80% for non-Newtonian flows. For shear-thinning fluids, the wall-shear history was reshaped, with a brief early amplification and late-phase mean reductions of 10.3% and 13.3% at ΔP1 and ΔP2. These results establish a fluid-mechanical mechanism linking clinically relevant vibration amplitudes to reduced sustained shear exposure, deeper and broader depot formation, and improved conditions for drug uptake.

The purpose of this research is to determine whether retinal vasculatures differ between participants with schizophrenia spectrum disorder (SSD) and controls. Ninety participants (51 SSD, mean age 35.8 ± 13.5, and 39 controls, mean age 35.5 ± 11.4) underwent 3 × 3 mm² macular and 6 × 6 mm² optic nerve head (ONH) optical coherence tomography angiography (OCTA) scans. En face macula and ONH region images were divided into quadrants, binarized, and then skeletonized. Skeletonized vessel densities were compared between our two groups. Additionally, the foveal avascular zone (FAZ) size and acircularity index were compared between the two groups. There was significantly decreased vessel density in the temporal region of the ONH in the SSD group compared to controls (p = 0.033). Interestingly, the decreased vessel density was already present in patients with SSD in younger adulthood as compared to the controls (p = 0.006). There were no significant group differences in vessel density in any other region of the ONH, the ONH overall, any region of the macula, or the macula overall. There were also no significant group differences in the FAZ size or acircularity index. These data suggest there may be abnormal peripapillary retinal vasculature in patients with SSD. Whether this is a specific ocular vascular deficit or related to more systemic vascular abnormalities in SSD remains to be determined.

Tricalcium phosphate (TCP) bioceramics, available as α- and β-polymorphs, are frequently employed in the production of three-dimensionally (3D) printed bone scaffolds. Although hydrothermal immersion processing (HP) and sintering (S) are commonly adopted as post-printing techniques for bioceramics, a comprehensive comparative analysis of their effects on the osteogenic performance of α- and β-polymorphs in vivo remains inadequately investigated. In this study, α-TCP and β-TCP scaffolds were fabricated via direct ink writing and subjected to hydrothermal immersion processing (α-TCP/HP) and sintering (β-TCP/S) prior to implantation in n = 12 skeletally mature sheep (n = 1 scaffold per group per animal), and the outcome variables were evaluated at 3 and 12 weeks postoperatively (n = 6 sheep per time point). The quantitative results showed no significant differences in bone deposition or scaffold resorption at 3 weeks postoperatively (p = 0.618 and p = 0.898, respectively). However, at 12 weeks, there was a significant increase in osteogenesis and scaffold resorption in the β-TCP/S cohort relative to the α-TCP/HP counterparts (p < 0.001 and p = 0.004, respectively). β-TCP scaffolds subjected to post-print sintering exhibited superior osteoconductive and resorptive profiles compared to hydrothermal immersion-processed α-TCP scaffolds over the 12-week healing period.

This study investigated methane (CH₄) production in a bioelectrochemically enhanced anaerobic digester (BEAD) equipped with a pair of 3-dimensional flow-through electrodes made of conductive polypropylene biorings. The performance of the BEAD reactor was compared to that of a similarly sized Anaerobic Upflow Sludge Bed (UASB) reactor. The reactors were operated at a temperature of 22 ± 1 °C using food waste (FW) leachate fed at organic loading rates of 3-8 g (L_R d)^-1 or at a temperature of 35 ± 1 °C using the liquid fraction of FW separated using a screw press. With both tested feedstocks, the BEAD reactor demonstrated up to 30% higher CH₄ yield, reaching 0.35-0.38 L g^-1 (COD consumed), compared to the UASB reactor. Additionally, reactor stability under organic overload conditions improved, with the difference more pronounced at organic loads above 6 g (L_R d)^-1. Energy consumption for bioelectrochemical CH₄ production was estimated at 5.1-12.4 Wh L^-1 (of CH₄ produced), which is significantly below the energy consumption for electrochemical H₂-based methanation. Overall, BEAD increases methane production and improves process stability, offering a novel sustainable solution for waste management.

Acoustic voice analysis demonstrates potential as a non-invasive biomarker for depression, yet its generalizability across languages remains underexplored. This cross-sectional study aimed to identify a set of cross-culturally consistent acoustic features for depression screening using distinct Vietnamese and Japanese voice datasets. We analyzed anonymized recordings from 251 participants, comprising 123 Vietnamese individuals assessed via the self-report Beck Depression Inventory (BDI) and 128 Japanese individuals assessed via the clinician-rated Hamilton Depression Rating Scale (HAM-D). From 6373 features extracted with openSMILE, a multi-stage selection pipeline identified 12 cross-cultural features, primarily from the auditory spectrum (AudSpec), Mel-Frequency Cepstral Coefficients (MFCCs), and logarithmic Harmonics-to-Noise Ratio (logHNR) domains. The cross-cultural model achieved a combined Area Under the Curve (AUC) of 0.934, with performance disparities observed between the Japanese (AUC = 0.993) and Vietnamese (AUC = 0.913) cohorts. This disparity may be attributed to dataset heterogeneity, including mismatched diagnostic tools and differing sample compositions (clinical vs. mixed community). Furthermore, the limited number of high-risk cases (n = 33) warrants cautious interpretation regarding the reliability of reported AUC values for severe depression classification. These findings suggest the presence of a core acoustic signature related to physiological psychomotor changes that may transcend linguistic boundaries. This study advances the exploration of global vocal biomarkers but underscores the need for prospective, standardized multilingual trials to overcome the limitations of secondary data analysis.

Background: Anatomopathological reports are largely unstructured, which limits automated data extraction, interoperability, and large-scale research. Manual extraction and standardization are costly and difficult to scale.

Objective: We developed and evaluated an automated pipeline for entity extraction and multi-ontology normalization of anatomopathological reports.

Methods: A corpus of 560 reports from the Military Hospital of Tunis, Tunisia, was manually annotated for three entity types: sample type, test performed, and finding. The entity extraction utilized BioBERT v1.1, while the normalization combined BioClinicalBERT multi-label classification with retrieval-augmented generation, incorporating both dense and BM25 sparse retrieval over SNOMED CT, LOINC, and ICD-11. The performance was measured using precision, recall, F1-score, and statistical tests.

Results: BioBERT achieved high extraction performance (F1: 0.97 for the sample type, 0.98 for the test performed, and 0.93 for the finding; overall 0.963, 95% CI: 0.933-0.982), with low absolute errors. For terminology mapping, the combination of BioClinicalBERT and dense retrieval outperformed the standalone and BM25-based approaches (macro-F1: 0.6159 for SNOMED CT, 0.9294 for LOINC, and 0.7201 for ICD-11). Cohen's Kappa ranged from 0.7829 to 0.9773, indicating substantial to near-perfect agreement.

Conclusions: The pipeline provides robust automated extraction and multi-ontology coding of anatomopathological entities, supporting transformer-based named entity recognition with retrieval-augmented generation. However, given the limitations of this study, multi-institutional validation is needed before clinical deployment.

We propose a novel continual self-supervised learning (CSSL) framework for simultaneously learning diverse features from multi-window-obtained chest computed tomography (CT) images and ensuring data privacy. Achieving a robust and highly generalizable model in medical image diagnosis is challenging, mainly because of issues, such as the scarcity of large-scale, accurately annotated datasets and domain shifts inherent to dynamic healthcare environments. Specifically, in chest CT, these domain shifts often arise from differences in window settings, which are optimized for distinct clinical purposes. Previous CSSL frameworks often mitigated domain shift by reusing past data, a typically impractical approach owing to privacy constraints. Our approach addresses these challenges by effectively capturing the relationship between previously learned knowledge and new information across different training stages through continual pretraining on unlabeled images. Specifically, by incorporating a latent replay-based mechanism into CSSL, our method mitigates catastrophic forgetting due to domain shifts during continual pretraining while ensuring data privacy. Additionally, we introduce a feature distillation technique that integrates Wasserstein distance-based knowledge distillation and batch-knowledge ensemble, enhancing the ability of the model to learn meaningful, domain-shift-robust representations. Finally, we validate our approach using chest CT images obtained across two different window settings, demonstrating superior performance compared with other approaches.

Biomaterials made with extracellular matrix obtained from allogeneic or xenogeneic tissues/organs or cultured cells have excellent biochemical and physical properties in supporting cell growth and tissue regeneration. These decellularized extracellular matrix-based biomaterials have been applied in clinical trials and have bright prospects in tissue engineering and regenerative medicine. Here, we systematically compare organ-derived and cell-derived decellularized extracellular matrix, summarize commonly used decellularization methods, including physical, chemical, and biological/enzymatic treatments, as well as combinations of these treatments, and characterize methods for decellularization, including histological staining, immunohistochemical techniques, biochemical analysis, scanning electron microscopy, and mechanical stress testing. Besides the production of decellularized extracellular matrix, the evolving intellectual property landscape and commercial products are also introduced. A significant focus is placed on summarizing clinical trial outcomes, demonstrating the efficacy of decellularized extracellular matrix scaffolds in diverse applications, including wound healing, cardiovascular repair, nerve regeneration, and breast reconstruction. Finally, we discuss persistent challenges and future directions, underscoring the translational potential of decellularized extracellular-matrix-based strategies for restoring tissue structure and function.

Emotion recognition based on EEG signals remains a challenging task due to the complex spatiotemporal properties of brain activity and substantial intersubject variability. To address these challenges, we propose the EED-CL framework, which integrates an extended EEG-Deformer (EED) with contrastive learning (CL). The proposed model incorporates a depthwise separable convolution encoder for efficient extraction of spatiotemporal EEG features, a hierarchical coarse-to-medium-to-fine (HCMFT) transformer to capture multiscale temporal patterns, and an attentive dense information purification (ADIP) module to suppress noise and refine essential latent representations. In addition, CL-based pretraining facilitates robust feature learning even in settings with limited labeled data. The extracted multiscale features are integrated and classified through a Transformer encoder and an MLP. Experiments conducted on multiple benchmark EEG datasets show that EED consistently outperforms conventional models, while EED-CL achieves further improvements under label-constrained conditions. Notably, EED-CL demonstrates strong robustness to intersubject variability and noise, enabling stable emotion classification even when labeled samples are scarce. These findings indicate that EED-CL effectively captures multiscale spatiotemporal EEG patterns and offers a scalable and reliable approach for EEG-based emotion recognition.

Focused ultrasound (FU) technology is extensively employed in clinical applications such as tumor ablation, Parkinson's disease treatment, and neuropathic pain management. The safety and efficacy of FU therapy critically depend on the accurate quantification of the acoustic field, particularly the high-pressure distribution in focal region. To address the limitations of existing acoustic measurement techniques-including invasiveness, inability to measure high sound pressure, and system complexity-this study proposes a non-invasive method termed Laser Deflection Acoustic Field Quantification (LDAQ), based on the laser deflection principle. An experimental system was constructed utilizing the acousto-optic deflection effect, which incorporates precision displacement control, rotational scanning, and synchronized triggering. Through tomographic scanning, laser deflection images of the acoustic field were acquired at multiple orientations. An inversion algorithm using Radon transforms was proposed to reconstruct the refractive index gradient distributions from the variations of light intensity and spot displacement. An adaptive weighted fusion strategy was then employed to map these optical signals to the sound pressure field. To validate the LDAQ technique, an acoustic field generated by an FU transducer operating at 0.84 MHz was measured. The reconstructed results were compared with both hydrophone measurements and numerical simulations. The findings demonstrated high consistency among all three results within the focal zone. Full-field analysis yielded a root mean square error (RMSE) of 0.1102 between LDAQ and simulation, and an RMSE of 0.1422 between LDAQ and hydrophone measurements. These results confirm that LDAQ enables non-invasive and high-precision quantification of megapascal-level focused acoustic fields, offering a reliable methodology for acoustic field characterization to support FU treatment optimization and device standardization.