Bioengineering最新文献

Background: Numerous protocols exist concerning the decellularization of the esophagus, a potential alternative to the classical surgical approach for the reconstruction of the digestive tract after esophagectomy. This systematic literature review (SLR) aimed to provide an overview of the effectiveness of the current protocols.

Methods: This SLR was conducted in PubMed, EMBASE, and Scopus until September 2025. Study selection, data extraction, and quality assessment were performed by two independent reviewers according to the inclusion/exclusion criteria.

Results: A total of 2494 references were screened after removing duplicates. Among these references, 26 articles were included. The large majority of studies (24/26) used Sodium Dodecyl Sulfate (SDS) or Sodium DeoxyCholate (SDC), and the most common physical method was the cannulation of the esophagus (17/26). The animal model was very heterogenous. All protocols except one showed no residual cell nuclei, with only 5/19 papers confirming a satisfactory residual amount of DNA. The assessment of the extracellular matrix (ECM)-mostly qualitative-revealed global preservation but with a systematic loss of glycosaminoglycans (GAGs).

Conclusions: The decellularization of the esophagus is feasible, but the definition of the optimal protocol to achieve this goal remains difficult because of the important heterogeneity among the different studies.

In this study, two native microalgae, Chlorella sp. MC18 (CH) and Scenedesmus sp. MJ23-R (SC) were cultivated in bubble column photobioreactors for wastewater treatment. Domestic wastewater (DWW) was used as the main culture medium, alone (100%) and blended (10%) with vinasse, whey, or agro-food waste (AFW), respectively. Both species thrived in 100% DWW, achieving significantly high removal efficiencies for chemical oxygen demand, total nitrogen, and total phosphorus. Mineral removal exceeded 90% in all blended systems, highlighting the strong nutrient uptake capacity of both strains. The maximum specific growth rate (µ_max) in 100% DWW was higher for SC than in standard BG11 medium, and supplementation with vinasse, whey, or AFW further increased µ_max for both species. Blending DWW significantly enhanced microalgal biomass and lipid production compared to 100% DWW. Lipid production (max., 374 mg L^-1), proximate lipid composition (max., 30.4%), and lipid productivity (max., 52.9 mg L^-1 d^-1) significantly increased in all supplemented cultures relative to DWW alone, demonstrating the potential of co-substrate supplementation to optimize microalgal cultivation. This study contributes to reducing the water footprint and fills a gap in the bioprocessing potential of algae-based systems, highlighting wastewater blending as a circular economy-aligned approach that supports sustainable bioprocesses and resource recovery.

Background: Domain-specific large language models (LLMs) like Ortho GPT have potential advantages over general-purpose models in medical education, offering improved factual accuracy and contextual relevance. This study evaluates the performance of Ortho GPT against general LLMs and senior medical students on validated orthopedic examination questions.

Methods: Six LLMs (Ortho GPT 4o, ChatGPT 4o, ChatGPT 3.5, Perplexity AI, DeepSeek-R1, and Llama 3.3-70B) were tested using multiple-choice items from final-year medical student orthopedic exams in German language. Each model answered identical questions under standardized zero-shot conditions; accuracy rates and item-level results were compared using McNemar's test, Jaccard similarity, and point-biserial correlation with student difficulty ratings.

Results: Ortho GPT achieved the highest accuracy across models. McNemar's tests revealed the significant superiority of Ortho GPT over DeepSeek (p = 2.33 × 10^-35), Llama 3.3-70B (p = 1.11 × 10^-32), and Perplexity (p = 4.01 × 10^-5). Differences between Ortho GPT and ChatGPT 4o were non-significant (p = 0.065), suggesting near-equivalent performance to the strongest general model. No LLM showed correlation with student item difficulty (|r| < 0.07, p > 0.05), indicating that models solved items independently of human-perceived difficulty. Jaccard indices suggested moderate overlap between Ortho GPT and ChatGPT 4o, but distinct response profiles compared with general LLMs.

Conclusions: These findings illustrate the superiority of Ortho GPT in orthopedic exam accuracy and context relevance, attributed to its specialized training data. The domain-specific approach enables performance matching or exceeding top general LLMs in orthopedics, emphasizing the importance of domain specialization for reliable, curriculum-aligned support in medical education.

Background and Objectives: Spirometry is the most widely used pulmonary function test for diagnosing respiratory diseases. Its progressive incorporation into non-specialized settings, such as primary care, raises challenges for ensuring the reliability of results. In this context, tools based on artificial intelligence (AI) techniques have emerged as promising solutions to support quality control in spirometry. This systematic review aims to synthesize the available evidence on their application in this field. Methods: A systematic search was conducted in PubMed and IEEE Xplore to identify peer-reviewed original studies, published between 2014 and June 2025, that applied AI to spirometry quality control. The search and data extraction followed the PRISMA guidelines. Results: Six studies met the inclusion criteria. Four analyzed the acceptability and usability of the maneuver, and two focused on detecting errors committed during test performance. The most widely used models were convolutional neural networks, used in four studies, whereas two studies employed other conventional machine learning models. Three models reported area under the ROC curve values higher than 0.88. Conclusions: AI-based tools show great potential to assist in spirometry quality control, both in determining acceptability and in detecting errors. However, current studies remain scarce and highly heterogeneous in both objectives and methods. Broader, multicenter research, including validation in non-specialized settings, is required to confirm their clinical utility and facilitate their implementation in clinical practice.

The objective of this study is to develop a robust diabetes diagnosis model by employing four distinct algorithms as weak learners: the Random Forest algorithm, SVM algorithm, KNN algorithm, and Decision Tree algorithm. The selection of the optimal classification model involves a meticulous process, and further refinement is conducted through the application of the Stacking classifier, with the Multilayer Perceptron (MLP) classifier serving as the final model. The performance of the optimized model is thoroughly evaluated to identify the most effective diagnostic model. The experiments are conducted using a dataset obtained from Changhua Christian Hospital in Taiwan. Our experimental results show that the performance of the model optimized by the Stacking ensemble learning method is significantly improved. The optimized model achieves an F₁ score of 0.86 and an AUC score of 0.90, indicating the effectiveness of the proposed model.

The increasing prevalence of orthopedic disorders and technological advances have significantly improved the design and functionality of orthopedic implants, fostering the growth of the orthopedic implant market. Polyetheretherketone (PEEK) has emerged as a promising alternative to the gold standard of metallic implants due to its favorable biocompatibility and mechanical properties, comparable to those of bone tissue. However, its chemical inertness results in poor osseointegration. This study investigates femtosecond (fs) laser-induced micro- and nanoscale surface modifications of PEEK, aiming to develop surface modifications potentially favorable for bioactivity enhancement of the as-created transient cellular scaffolds. Various texturing designs were fabricated by precisely controlling the laser parameters applied (laser beam power P = 20-80 mW, hatch spacing dx = 45-100 µm, scanning velocity V = 3.44-32 mm/s). The resulting morphologies were characterized by SEM, EDX, XRD, micro-Raman, 3D profilometry, water contact angle measurements, and evaluated for preliminary biological response. The main achievement of the research indicates that the hierarchical topography created by fs laser microprocessing significantly increased surface morphology, which may subsequently provide surface conditions supporting successful osseointegration. These findings demonstrate the feasibility of femtosecond laser structuring as a promising, reproducible, and environmentally friendly method for sustainable surface biofunctionalization of PEEK in orthopedic applications.

Background: Radiomic feature robustness is a key prerequisite for the reproducibility and clinical translation of imaging biomarkers. Variability across software platforms can significantly affect feature consistency, compromising predictive modeling reliability. This study aimed to develop and validate a hierarchical clustering-based workflow for evaluating radiomic feature robustness within and across software platforms, identifying stable and reproducible features suitable for clinical applications. Methods: A multi-cancer CT dataset including 97 lesions from 71 patients, comprising primary colorectal cancer (CRC), colorectal liver metastases, and hepatocellular carcinoma (HCC), was analyzed. Radiomic features were extracted using two IBSI-compliant platforms (MM Radiomics of syngo.via Frontier and 3D Slicer with PyRadiomics). Intra-software reliability was assessed through the intraclass correlation coefficient ICC(A,1), while cross-software stability was evaluated using hierarchical clustering validated by the Adjusted Rand Index (ARI). A Composite Index (CI) integrating correlation, distributional similarity, and mean fractional ratio quantified inter-platform feature robustness. Results: Over 95% of radiomic features demonstrated good-to-excellent intra-software reliability. Several clustering configurations achieved ARI = 1.0, confirming strong cross-platform concordance. The most robust and recurrent features were predominantly wavelet-derived descriptors and first-order statistics, particularly cluster shade (GLCM-based) and mean intensity-related features. Conclusions: The proposed multi-stage framework effectively identifies stable, non-redundant, and transferable radiomic features across IBSI-compliant software platforms. These findings provide a methodological foundation for cross-platform harmonization and enhance the reproducibility of radiomic biomarkers in oncologic imaging.

Slips are a leading cause of injury and hospitalization among at-risk individuals. Replicating real-world slips by experimental, mechanical perturbations is essential for characterizing reactive balance mechanisms activated during near-fall situations and for training these mechanisms in fall prevention programs. This study compared treadmill-based, slip-like perturbations targeting the early (early perturbations, EP) vs. late stance phase (late perturbations, LP) in 22 young and 21 older adults. Biomechanical and neuromuscular responses were assessed using full-body kinematics and surface electromyography (EMG). Additionally, participants provided subjective rating of perturbation intensity and inconvenience. EP elicited stronger reactive balance responses than LP, characterized by greater deviations in leg joint and trunk kinematics, as well as shorter EMG onset latencies and enhanced EMG peak amplitudes. Gait parameters required longer to recover to baseline following EP than LP. Subjectively, EP were rated as more intense and inconvenient, and were perceived to more closely mimic real-world slips. Older adults showed delayed and attenuated reactive balance responses compared to younger adults. These findings highlight the importance of targeting the vulnerable early stance phase to accurately simulate real-world slip events. Such perturbation paradigms may support the development of more effective, task-specific perturbation-based training programs aimed at reducing falls in at-risk populations.

The widely used oximeter design was adopted and improved as the configuration mainframe in this study to acquire PPG signals. When users wear a finger probe and input their height, the device acquires PPG signals through the probe circuit, then filters and amplifies the signals to remove unnecessary noise, and uses an ARM-M4 to analyze the main peak, dicrotic wave, and wave valley of the PPG waveform to calculate related indexes for the final assessment. After 100 s, the HRV, sine wave ratio, and SI results are estimated, and a cardiovascular disease risk assessment is presented using a risk level from 0 to 5. This study uses the stiffness index (SI), sine wave ratio (SIN), and heart rate variability (HRV) to assess cardiovascular status. The SI is derived from PPG signal characteristics and reflects vascular stiffness based on blood flow rebound time. However, some PPG signals lack a dicrotic wave (sine waves), which is often caused by severe arterial stiffness. These waveforms lead to errors in SI calculation due to misidentification of the dicrotic wave. The appearance of a sine wave indicates that blood pulsation is abnormal; however, it will make the SI calculation algorithm produce a seemingly normal health performance. To address this, the auxiliary line method was introduced to identify sine waves, and the SIN ratio occurring in contiguous PPG waves was incorporated to calculate their proportion in PPG signals, aiding SI analysis and arterial stiffness evaluation. The total power (TP) value obtained via HRV frequency-domain analysis reflects autonomic nervous activity. As reduced autonomic function may relate to cardiovascular diseases, TP is included as an evaluation indicator. By analyzing PPG signals, calculating SI and SIN, and integrating the HRV indicator, this study evaluates arterial stiffness and cardiovascular health, helping participants understand their physical condition more quickly and conveniently, and potentially preventing cardiovascular diseases at an early stage.

(1) Background: Anterior cruciate ligament reconstruction (ACLR) alters lower-limb biomechanics. While gait and running are well-studied, the multi-phase side-cutting remains poorly understood, particularly regarding phase-specific adaptations after ACLR. (2) Methods: Thirty-four patients (19 male, 15 female) at nine months post-ACLR participated. Biomechanical data during side-cutting were collected using synchronized motion capture and force platforms. Knee joint kinematics and kinetics were analyzed over three phases: initial contact-deceleration, stance pivot, and push-off. (3) Results: During the initial contact-deceleration, the reconstructed limb exhibited greater knee external rotation at the first posterior ground reaction force (pGRF) peak (8.5° vs. 6.3°, p = 0.021), and reduced knee flexion (43.2° vs. 47.3°, p < 0.001) with a lower extension moment at the second pGRF peak (0.10 vs. 0.14 BW·BH; p < 0.001). The stance pivot phase was marked by significantly lower knee flexion (p = 0.001), extension moment (p < 0.001), and medial/vertical GRFs on the reconstructed side (0.49 vs. 0.52 BW, p = 0.029; 1.98 vs. 2.10 BW, p = 0.012). During the push-off, the involved limb demonstrated a significantly lower extension moment (0.008 vs. 0.014 BW·BH, p = 0.037) and anterior GRF (0.20 vs. 0.23 BW, p = 0.010). (4) Conclusions: This study proposes a three-phase compensation model for side-cutting: "rotational instability" at initial contact, "protective unloading" during the stance pivot phase, and "force-generation deficit" at push-off. This three-phase framework provides a new paradigm for evaluating dynamic knee function after ACLR and guiding phase-specific rehabilitation.