Bioengineering MDPI

29/08/2026

💥Excited for the publication: "Machine Learning Integrating SERF-MEG and VEP for Diagnosis and Differential Diagnosis of Optic Neuropathies"
🔗https://brnw.ch/21x5iCp
💡Visual evoked potentials are widely used to assess optic nerve function, but distinguishing between different optic neuropathies remains difficult. This study evaluated spin-exchange relaxation-free magnetoencephalography together with machine learning to improve diagnostic classification of optic neuritis and ischemic optic neuropathy. Combining VEP and MEG features performed best for separating healthy eyes from optic neuropathies, while MEG alone was more effective for distinguishing between the two disease types. The findings suggest that SERF-MEG can provide complementary neurophysiological information beyond conventional VEP testing.
📌

28/08/2026

🧬 Explore the Bioengineering 2025 Second Half-Year Issue Cover Papers, a curated collection of exceptional research showcasing innovation, interdisciplinary excellence, and impactful breakthroughs shaping the future of bioengineering.

🔬 Aerosol Jet Printing for Neuroprosthetic Device Development
🔗 https://brnw.ch/21x5ief

🔬 Generation of an In Vitro Cartilage Aging Model Using Human Sera from Old Donors
🔗 https://brnw.ch/21x5ieg

🔬 Porous Structures, Surface Modifications, and Smart Technologies for Total Ankle Arthroplasty: A Narrative Review
🔗 https://brnw.ch/21x5iei

🔬 Electrospun Polycaprolactone/Collagen Scaffolds Enhance Manipulability and Influence the Composition of Self-Assembled Extracellular Matrix
🔗 https://brnw.ch/21x5iee

🔬 Capturing Compensatory Reserve in Sarcopenia: A Bioengineering Framework for Multidimensional Temporal Analysis of Center-of-Pressure Signals
🔗 https://brnw.ch/21x5ieh

🔬 A Comprehensive Review: The Bidirectional Role of Sebum in Skin Health
🔗 https://brnw.ch/21x5ied

The publications can also be accessed via the QRs below.

28/08/2026

💥Excited for the publication: "Recent Advances in Lipid Nanoparticle-Mediated Respiratory and Gastrointestinal Mucosal Delivery of Nucleic Acids"
🔗https://brnw.ch/21x5i4u
💡Delivering nucleic acid therapeutics across respiratory and gastrointestinal mucosa remains challenging because of mucus barriers, poor cellular uptake, endosomal trapping, and rapid clearance. This review examines how lipid nanoparticles can be engineered to overcome these obstacles through control of particle size, surface charge, deformability, and PEGylation. It also explores how future delivery systems could combine mucosa-specific physiological information with artificial intelligence to develop more adaptive and personalized nucleic acid therapies.
📌

28/08/2026

🦿 Standard gait analysis tells us a lot about how we move — but not which muscles are doing the work. A narrative review in Bioengineering explores the tools researchers use to go deeper, from ultrasound imaging to computational muscle modelling.

🥼 Authored by Stephen J. Piazza.

🔬 These methods are helping clinicians make better decisions about surgeries like tendon lengthening, especially for patients with cerebral palsy, Parkinson's disease, and stroke.

You may access the full article freely here 👉 https://brnw.ch/21x5hyG
Please feel free to follow our Facebook account Bioengineering MDPI!

Three-dimensional motion analysis performed in the modern gait analysis laboratory provides a wealth of information about the kinematics and kinetics of human locomotion, but standard gait analysis is largely restricted to joint-level measures.

28/08/2026

💥Excited for the publication: "Combined Transcranial Direct Current Stimulation and Virtual Reality in Healthy Populations: A Systematic Review of Evidence, Limitations, and Methodological Challenges"
🔗https://brnw.ch/21x5h2J
💡Combining brain stimulation with virtual reality could offer new ways to enhance learning and behavioral training, but how much does tDCS actually add to VR-based interventions? This systematic review examined 12 studies combining transcranial direct current stimulation with virtual reality in healthy populations. Some evidence suggests that active tDCS may provide additional benefits for emotional control and skill acquisition, but findings across cognitive outcomes remain inconsistent. Current evidence does not yet establish a synergistic effect between tDCS and VR, emphasizing the need for larger, well-controlled studies that can separate their individual and combined effects.
📌

28/08/2026

🧫 Training a reliable AI sensor on just 11 fermentation experiments sounds like a recipe for failure. A VAE that generates 100 plausible new experiments from those 11 turns that limitation into an advantage.

🥼 Authored by Hyun J. Kwon, Joseph H. Shiu, Celina K. Yamakawa, and Elmer C. Rivera.

🔬 The VAE-generated synthetic data not only preserved the statistical distribution of the original fermentation time series but also filled the solution space between experiments, including bridging the outlier caused by a miscalibrated capacitance sensor and the normal cluster. The augmented soft sensor handled real-world sensor noise far better than the original model, demonstrating practical robustness for large-scale bioprocess monitoring.

You may access the full article freely here 👉 https://brnw.ch/21x5gOS

Please feel free to follow our Facebook account Bioengineering MDPI!

Soft sensors based on deep learning regression models are promising approaches to predict real-time fermentation process quality measurements. However, experimental datasets are generally sparse and may contain outliers or corrupted data. This leads to insufficient model prediction performance. Ther...

28/08/2026

🤝 🧬Meet Us at the BMES 2026 Annual Meeting, 21–24 October 2026, Orlando, USA 🇺🇸

🔬Conference: BMES 2026 Annual Meeting
⚙️Organization: Biomedical Engineering Society
📆Date: 21–24 October 2026
🏙️Place: Orlando, USA

If you are attending the BMES 2026 Annual Meeting, we invite you to visit us at our booth #1302. 📍👋 Our representatives will be available to discuss publishing opportunities, the benefits of open access, and answer any questions that you may have. 📄💡❓

More info 🔗 https://brnw.ch/21x5gC2

Conference: BMES 2026 Annual Meeting Organization: Biomedical Engineering Society Date: 21–24 October 2026 Place: Orlando, USA
We are delighted to...

28/08/2026

🦷 Deciding whether to extract permanent teeth before orthodontics involves weighing dozens of clinical and radiographic variables, and two orthodontists presented with the same patient can reach different conclusions. AI trained on real-world multi-clinician data can begin to systematize that judgment.

🥼 Authored by Lily E. Etemad, J. Parker Heiner, A. A. Amin, Tai-Hsien Wu, Wei-Lun Chao, Shin-Jung Hsieh, Zongyang Sun, Camille Guez, and Ching-Chang Ko.

🔬 This multi-institutional study found that combining data from two US university clinics, despite different treatment philosophies and patient populations, improved random forest model performance to 85% accuracy. Maxillary and mandibular crowding were the dominant predictors at both institutions, while incisor position variables (U1-NA at OSU; L1-NB and FMIA at UNC) differed, reflecting institutional treatment philosophy differences rather than population variation.

You may access the full article freely here 👉 https://brnw.ch/21x5gxe

Please feel free to follow our Facebook account Bioengineering MDPI!

The study aimed to evaluate the effectiveness of machine learning in predicting whether orthodontic patients would require extraction or non-extraction treatment using data from two university datasets. A total of 1135 patients, with 297 from University 1 and 838 from University 2, were included dur...

27/08/2026

💥Excited for the publication: "Generalized Retinal Artery/Vein Segmentation via Multi-Dataset Fine-Tuning and Pathology Subgroup Analysis"
🔗https://brnw.ch/21x5gea
💡Accurate artery and vein segmentation in retinal images is important for deriving vascular biomarkers, yet models trained on individual datasets often struggle with heterogeneous imaging conditions and disease presentations. This study jointly fine-tuned a single deep learning model across eight fundus datasets using a loss strategy that accommodates both artery/vein-labeled and vessel-only images. The resulting model showed consistent segmentation performance across multiple datasets and retinal pathologies, including AMD, glaucoma, and diabetic retinopathy. While these findings demonstrate promising internal consistency, external patient-level validation is still needed to establish true cross-domain robustness.
📌

27/08/2026

🔊 Every ultrasound frequency involves a trade-off: higher frequencies give better resolution but more noise, while lower frequencies pe*****te deeper but with less detail. Deep learning may resolve both sides of this compromise more effectively than conventional filters.

🥼 Authored by Hyekyoung Kang, Chanrok Park, and Hyungjin Yang.

🔬 This study directly compared a ResNet-based denoiser against median, Wiener, and MMWF filters across clinical ultrasound frequencies (3 and 5 MHz) at two noise intensities. The ResNet consistently delivered higher PSNR, lower RMSE, and better-preserved edge detail, with the largest advantage appearing at higher noise levels (σ=0.1), exactly the conditions where conventional filters degrade most.

You may access the full article freely here 👉 https://brnw.ch/21x5fEB
Please feel free to follow our Facebook account Bioengineering MDPI

Ultrasound imaging is widely used for accurate diagnosis due to its noninvasive nature and the absence of radiation exposure, which is achieved by controlling the scan frequency. In addition, Gaussian and speckle noises degrade image quality. To address this issue, filtering techniques are typically...