BioMedical Engineering OnLine最新文献

During the past decades, three-dimensional (3D) printing processes have come as the foremost technology for the fabrication of scaffolds in tissue engineering (TE). The advanced technical approaches followed by 3D printing have provided architectural versatility and customizability. Despite the many progresses, several limitations have emerged related to the available, processable range of materials offering desired functions equivalent or suitable for the target tissue. To address the issue raised, several novel methodologies have been developed where a 3D printed sacrificial mold serves to produce the final scaffold from a wide range of materials, even from the difficult-to-print or unprintable materials. These techniques are known as "indirect 3D printing" (I3DP), which like the direct 3D printing approaches, are able to manufacture controlled, patient-specific constructs. Direct 3D printing faces limitations like poor printability of natural soft polymers and bio-ceramics, restricted resolution of the printed objects, and a limited range of compatible materials. Indirect 3D printing overcomes these by enabling the use of a much wider variety of materials and creating high-strength ceramic scaffolds without clogging or structural defects. This method also provides superior resolution with less parameter optimization and minimizes material waste, making it more efficient. The current review paper presents a state-of-the-art study of how indirect 3D printing is being utilized in tissue engineering. The focus is given to the details of steps required for the production of scaffolds including mold design, software, 3D printing machines used, mold and scaffold materials, mold removal approaches, combination with other pore forming methodologies and the area of applications in tissue engineering.

Background: The aim of this study was to investigate changes in corneal biomechanical properties after femtosecond laser in situ keratomileusis (FS-LASIK) and orthokeratology (Ortho-K).

Methods: In this retrospective comparative study, patients were divided into an Ortho-K group (37 eyes) and a LASIK group (35 eyes) based on whether they underwent Ortho-K correction or FS-LASIK surgery. Keratometry measurement (Km), spherical equivalent (SE), non-contact tonometry (NCT), central corneal thickness (CCT), corneal hysteresis (CH), and corneal resistance factor (CRF) were assessed at baseline and 1 month after treatment (Ortho-K or LASIK).

Results: No significant baseline differences existed between the two groups. After correction, the LASIK group showed significantly lower values in Km, SE, CCT, NCT, CH, and CRF compared to the Ortho-K group (all P < 0.05). Intragroup comparisons revealed significant postoperative reductions in the LASIK group for CCT (from 543.37 ± 36.75 µm to 476.80 ± 38.75 µm, P < 0.001), CH (from 10.68 ± 1.46 mmHg to 9.03 ± 1.52 mmHg, P < 0.001), and CRF (from 10.18 ± 1.89 mmHg to 7.68 ± 1.89 mmHg, P < 0.001). In the Ortho-K group, CH and CRF decreased significantly (CH: 10.98 ± 1.43 to 9.81 ± 1.19 mmHg, P = 0.005; CRF: 10.77 ± 1.70 to 9.42 ± 1.52 mmHg, P < 0.001), but CCT change was insignificant. Analysis of ORA signals confirmed a greater biomechanical impact after LASIK, with significant reductions in both P1 and P2 (both P < 0.001), compared to only a P1 reduction in the Ortho-K group (P = 0.033).

Conclusions: Although both FS-LASIK and Ortho-K reduce CH and CRF values as measured by the ORA, their clinical implications are fundamentally different. The changes induced by FS-LASIK represent a true biomechanical weakening of the cornea, whereas the reductions observed after Ortho-K are primarily measurement artifacts caused by alterations in corneal curvature. Consequently, a more comprehensive approach is required for accurately assessing the corneal biomechanical status following Ortho-K.

Advances in 3D bioprinting technology are increasingly shaping medical applications, offering practical opportunities in tissue engineering, regenerative medicine, and personalized healthcare. By enabling the precise deposition of cells and biomaterials, 3D bioprinting allows the fabrication of functional, tissue-like constructs that reproduce key aspects of native human organs. Concrete progress has been demonstrated in applications, such as cartilage repair, skin grafts, and liver tissue models, which illustrate the translational potential of this technology. In addition, 3D bioprinted constructs are being explored for organ transplantation, drug testing, and disease modeling, where they can provide more physiologically relevant data than traditional models. Despite these advances, major challenges remain, including vascularization, mechanical stability, and ensuring long-term tissue functionality. The development of robust bioinks, regulatory acceptance, and the high cost of bioprinting platforms also represent significant barriers to widespread clinical adoption. This article reviews both the opportunities and challenges of 3D bioprinting in medicine, highlighting recent technological progress, ongoing preclinical research, and potential strategies for overcoming current limitations to accelerate clinical translation. Ultimately, 3D bioprinting is moving from proof-of-concept studies toward early clinical applications, underscoring its potential to become a transformative tool in regenerative medicine.

Background: Necrotizing enterocolitis (NEC) is an inflammatory intestinal disease that primarily affects premature infants and is a major cause of death in the neonatal period. Approximately half of the affected infants require surgical intervention, but there is no established consensus on the criteria or timing for surgery, making treatment decisions challenging.

Methods: 291 patients between 2019 and 2023 were categorized into medical group, early surgery group, and late surgery group. The region of interest (ROI) on the abdominal stereotaxic film underwent color-channel transformation, and radiomics features were subsequently extracted. Statistical analysis was then performed using a two-sample independent t test and least absolute shrinkage and selection operator (Lasso) regression to assess the relevant features.

Results: By screening and analyzing abdominal X-ray features across different subgroups, three key features-Blue_variance, a_star_variance, and L_star_energy-were identified as common differential markers. These features may serve as imaging biomarkers for predicting the surgical necessity in NEC patients.

Conclusion: We identified new parameters of abdominal X-rays in patients with NEC to assess the feasibility of surgical treatment and may provide supporting clinical guidance to clinicians.

Background: COVID-19 may impair autonomic and cardiorespiratory function, even in mild cases, resulting in reduced heart rate variability (HRV) and diminished functional capacity. Given their shared regulatory pathways, resting HRV may serve as a non-invasive predictor of oxygen uptake (V̇O₂) during the six-minute step test (6MST).

Objective: To investigate whether resting short-term HRV can predict V̇O₂ during the 6MST in individuals recovering from post-COVID.

Methods: In this cross-sectional study, adults recovering from mild COVID underwent assessment of autonomic modulation via short-term HRV and cardiorespiratory response during 6MST. HRV was recorded under standardized resting conditions. Gas exchange was measured throughout the 6MST. Spearman correlation and multiple linear regression analyses were performed to test associations between HRV parameters and V̇O₂ expressed in milliliters per kilogram per minute (mL·kg^-1·min^-1).

Results: Data from 45 participants were analyzed. Several HRV variables demonstrated statistically significant correlations with V̇O₂ and were therefore included in the simple linear regression analysis: SDNN (ms) (rho = 0.587), RMSSD (ms) (rho = 0.430), RR Tri (rho = 0.594), TINN (rho = 0.596), SD1 (ms) (rho = 0.431), SD2 (ms) (rho = 0.609), ApEn (rho = - 0.388), and DFA α2 (rho = - 0.404). Multiple linear regression showed that SD2 (ms) and sex were significant predictors of V̇O₂ (mL·kg^-1·min^-1) at the peak of the 6MST, while weight (kg) and age (years) were not. The model explained 50.7% of the variance (adjusted R² = 0.507, p < 0.001).

Conclusion: Several HRV parameters were significantly correlated with V̇O₂, indicating associations between cardiac autonomic modulation and aerobic performance. Among these, SD2 together with sex, emerged as significant predictors of VO₂ at the peak of the 6MST. Future studies will be needed to combine HRV indices with clinical outcomes in order to determine the mechanisms of V̇O₂ variability in post-COVID populations.

Proprioceptive deficits have been linked to balance and risk of falling in older adults. In this study, we assessed the effects of proprioceptive deficits on sensorimotor performance using a sensor-based real-time tracking game (SRT) that utilizes a moving target on a screen. The current methods to measure proprioceptive deficits can be difficult to use and may not be sensitive to minor deficits. We recruited 19 young participants (mean age = 21.8 ± 1.32, 55% female) and 10 older adults (age = 73.6 ± 7.73, 60% female). We placed a gyroscope on the top of the intermediate cuneiform bone to track movements. Both legs were tested with three separate SRT tracks with different difficulties. To disturb visuo-proprioceptive performance, we applied vibration to the tibialis anterior, peroneus longus, soleus, and gastrocnemius. We hypothesized that younger participants would score better than older participants, the dominant ankle would perform better than the non-dominant ankle, and participants would perform better without vibratory stimulation. We measured the amplitude and directional accuracy of the performance. Amplitude accuracy was measured by calculating the straight-line distance from the tracker to predetermined track. Directional accuracy was defined as the percentage of time the participant remained in the free zone (defined by average performance among the young sample) at every instant of time. Results showed significant effects for amplitude and directional accuracy by age (p < 0.001), vibration condition (p < 0.012), and testing side (p < 0.026). Our results showed that SRT is sensitive enough to measure effects of visuo-proprioception changes caused by aging, dominant vs. non-dominant sides, and outside disturbance. Through accurate detection of visuo-proprioceptive deficits, we may potentially identify risk of falling associated with such deficits.

Background: Hemodialysis equipment alarms significantly impact clinical workflow and patient safety. This study aimed to characterize alarm patterns, identify risk factors, and evaluate management effectiveness in hemodialysis facilities.

Methods: A retrospective cohort analysis was conducted at 1 hemodialysis center from January 2020 to December 2023. Equipment alarm data from hemodialysis stations (Nikkiso, Gambro, and B.Braun) were systematically collected and analyzed. Multivariate regression and machine learning approaches identified risk factors and developed predictive model.

Results: Among 4231 recorded alarm events over 32 months, the monthly alarm rate averaged 132.2 ± 54.4 events. The COVID-19 pandemic period (2020-2021) accounted for 82.2% of alarms, with a significant 78.3% reduction post-pandemic (P = 0.022). B.Braun equipment generated 68.7% of alarms, significantly higher than Gambro (21.2%) and Nikkiso (10.1%) devices (P < 0.001). Engineering analysis revealed B.Braun's lower pressure thresholds (150 vs. 180 mmHg), higher sensor sensitivity (1 vs. 2-5 mmHg resolution), and conservative air detection algorithms (0.3 vs. 0.5 mL) contributed to increased alarm frequency. Pressure-related alarms predominated (37.4%), while 94.5% occurred during active treatment phases. Vascular access complications contributed to 19.8% of events, with permanent catheters being the primary source (47.3%). Patient-related factors accounted for 83.6% of human factor contributions. Patients with BMI > 30 kg/m² experienced 28% more alarms (OR 1.28, 95% CI 1.12-1.47, P < 0.001), highlighting the need for individualized alarm strategies. The alarm management protocol achieved a 99.3% immediate resolution rate. Predictive modeling (sensitivity 72.3%, specificity 75.8%, PPV 68.9%, NPV 78.6%) enabled targeted interventions, resulting in a 43% alarm reduction, a 56% false alarm decrease, and 51% fewer treatment interruptions (all P < 0.001).

Conclusions: Equipment-specific alarm patterns and human factors significantly influence hemodialysis alarm burden. Implementation of predictive analytics and targeted interventions substantially improves alarm management effectiveness and clinical outcomes.

Respiratory disease, such as lung cancer, is a major risk factor that seriously endangers human health. In the development of new drugs, conventional preclinical and clinical testing stages rely on cell culture and animal experiment. However, the two methods may fall short of fully representing human biology, thereby presenting an opportunity to advance technological innovation. In this review, we focus on the structure and the composition of supporting cells of organ-on-a-chip (OOAC), whose most notable feature is its biomimetic representation of the human body. Its core advantage lies in its biomimetic human structure, which enables broad application scenarios in the field of pulmonary diseases including lung cancer, pneumonia, pulmonary edema, and pulmonary embolism. Finally, this review summarizes the current challenges and prospects of OOAC, highlighting its vast potential for advancement.

Purpose: Fetal ultrasound imaging is critical for prenatal care, demanding accurate anatomical interpretation. This study evaluates the potential of Vision Large Language Models (LLMs) in interpreting fetal ultrasound images, exploring whether tailored prompts can enhance performance compared to standard prompts, and assessing their utility in clinical settings.

Methods: Nine fetal ultrasound images were analyzed using six advanced Vision LLMs via the Chatbot Arena platform. Standard prompts were compared against expert-crafted tailored questions. Three expert sonographers assessed the models' outputs across five criteria-anatomical recognition, biometric potential, picture quality, normalcy assessment, and clinical recommendations-using a Likert scale (1-5).

Results: Standard prompts yielded limited interpretative accuracy. In contrast, custom prompts significantly improved performance, with Claude Sonnet 3.5 and ChatGPT4o achieving median scores of 19 and 18, respectively. Models excelled in analyzing fetal femur and trans-cerebellum images, with clinical advice being the easiest to identify. Challenges persisted in precise anatomical identification and image quality assessment, revealing limitations in visual recognition. Smaller models like pixtral-12b showed notable improvement with tailoring, suggesting fine-tuning potential, while larger models did not consistently outperform smaller ones, indicating factors beyond model size influence efficacy.

Conclusion: Tailored prompts markedly enhance Vision LLMs' ability to interpret fetal ultrasound images, supporting their potential as aids in prenatal diagnosis and education. However, limitations in anatomical precision and image quality assessment persist. Future research should focus on refining models with specialized datasets, optimizing architectures, and advancing prompt engineering to maximize clinical utility.