Bioengineering

21 pages, 1594 KiB

Open AccessArticle

Implementation of a Conditional Latent Diffusion-Based Generative Model to Synthetically Create Unlabeled Histopathological Images

by Mahfujul Islam Rumman, Naoaki Ono, Kenoki Ohuchida, Ahmad Kamal Nasution, Muhammad Alqaaf, Md. Altaf-Ul-Amin and Shigehiko Kanaya

Bioengineering 2025, 12(7), 764; https://doi.org/10.3390/bioengineering12070764 (registering DOI) - 15 Jul 2025

Abstract

Generative image models have revolutionized artificial intelligence by enabling the synthesis of high-quality, realistic images. These models utilize deep learning techniques to learn complex data distributions and generate novel images that closely resemble the training dataset. Recent advancements, particularly in diffusion models, have [...] Read more.

Generative image models have revolutionized artificial intelligence by enabling the synthesis of high-quality, realistic images. These models utilize deep learning techniques to learn complex data distributions and generate novel images that closely resemble the training dataset. Recent advancements, particularly in diffusion models, have led to remarkable improvements in image fidelity, diversity, and controllability. In this work, we investigate the application of a conditional latent diffusion model in the healthcare domain. Specifically, we trained a latent diffusion model using unlabeled histopathology images. Initially, these images were embedded into a lower-dimensional latent space using a Vector Quantized Generative Adversarial Network (VQ-GAN). Subsequently, a diffusion process was applied within this latent space, and clustering was performed on the resulting latent features. The clustering results were then used as a conditioning mechanism for the diffusion model, enabling conditional image generation. Finally, we determined the optimal number of clusters using cluster validation metrics and assessed the quality of the synthetic images through quantitative methods. To enhance the interpretability of the synthetic image generation process, expert input was incorporated into the cluster assignments. Full article

(This article belongs to the Section Biosignal Processing)

15 pages, 3505 KiB

Open AccessArticle

Far-Red Component Enhances Paramylon Production in Photoautotrophic Euglena gracilis

by Zhaida I. Aguilar-Gonzalez, Anaiza Rico-Luna, Tóshiko Takahashi-Íñiguez and Héctor V. Miranda-Astudillo

Bioengineering 2025, 12(7), 763; https://doi.org/10.3390/bioengineering12070763 (registering DOI) - 15 Jul 2025

Abstract

In recent years, microalgae have gained significant biotechnological importance as a sustainable source of various metabolites of industrial interest. Among these, paramylon, a polysaccharide produced by the microalga Euglena gracilis, stands out for its diverse applications in biomedicine and pharmaceuticals. E. gracilis [...] Read more.

In recent years, microalgae have gained significant biotechnological importance as a sustainable source of various metabolites of industrial interest. Among these, paramylon, a polysaccharide produced by the microalga Euglena gracilis, stands out for its diverse applications in biomedicine and pharmaceuticals. E. gracilis is an adaptable secondary eukaryote capable of growing photoautotrophically, heterotrophically and mixotrophically. During photoautotrophic growth, varying light conditions impact biomass and paramylon production. To investigate the effects of varying illumination more thoroughly, we designed and built a modular photobioreactor that allowed us to simultaneously evaluate the photoautotrophic growth of E. gracilis under twelve different light conditions: seven single-spectrum lights (ultraviolet, royal blue, blue, green, red, far-red, and infrared) and five composite-spectrum lights (3000 K, 10,000 K, and 30,000 K white lights, amber light, and “Full-spectrum” light). The 24-day growing kinetics were recorded, and the growth parameters were calculated for each light regime. Both growth curves and pigment composition present differences attributable to the light regime used for cell culture. Additionally, photosynthetic and respiratory machinery functionality were proven by oximetry. Finally, our results strongly suggest that the far-red component enhances paramylon production during the stationary phase. Full article

(This article belongs to the Special Issue Microalgae Biotechnology and Microbiology: Prospects and Applications)

► Show Figures

Figure 1

11 pages, 872 KiB

Open AccessArticle

Assessment of Radiation Attenuation Properties in Dental Implants Using Monte Carlo Method

by Ali Rasat, Selmi Tunc, Yigit Ali Uncu and Hasan Ozdogan

Bioengineering 2025, 12(7), 762; https://doi.org/10.3390/bioengineering12070762 (registering DOI) - 14 Jul 2025

Abstract

This study investigated the radiation attenuation characteristics of commonly used dental implant materials across an energy spectrum relevant to dental radiology. Two titanium implants were examined, with densities of 4.428 g/cm³ and 4.51 g/cm³, respectively. The first consisted of 90.39% [...] Read more.

This study investigated the radiation attenuation characteristics of commonly used dental implant materials across an energy spectrum relevant to dental radiology. Two titanium implants were examined, with densities of 4.428 g/cm³ and 4.51 g/cm³, respectively. The first consisted of 90.39% titanium, 5.40% aluminum, and 4.21% vanadium, while the second comprised 58% titanium, 33% oxygen, 7% iron, 1% carbon, and 1% nitrogen. The third material was a zirconia implant (5Y form) composed of 94.75% zirconium dioxide, 5.00% yttrium oxide, and 0.25% aluminum oxide, exhibiting a higher density of 6.05 g/cm³. Monte Carlo simulations (MCNP6) and XCOM data were utilized to estimate photon source parameters, geometric configuration, and interactions with biological materials to calculate the half-value layer, mean free path, and tenth-value layer at varying photon energies. The results indicated that titanium alloys are well suited for low-energy imaging modalities such as CBCT and panoramic radiography due to their reduced artifact production. While zirconia implants demonstrated superior attenuation at higher energies (e.g., CT), their higher density may induce beam-hardening artifacts in low-energy systems. Future research should validate these simulation results through in vitro and clinical imaging and further explore the correlation between material-specific attenuation and CBCT image artifacts. Full article

(This article belongs to the Special Issue Functionalization of Dental Materials towards Improved Interaction with the Oral Environment)

► Show Figures

Figure 1

22 pages, 3456 KiB

Open AccessArticle

Precision in 3D: A Fast and Accurate Algorithm for Reproducible Motoneuron Structure and Protein Expression Analysis

by Morgan Highlander, Shelby Ward, Bradley LeHoty, Teresa Garrett and Sherif Elbasiouny

Bioengineering 2025, 12(7), 761; https://doi.org/10.3390/bioengineering12070761 (registering DOI) - 14 Jul 2025

Abstract

Structural analysis of motoneuron somas and their associated proteins via immunohistochemistry (IHC) remains tedious and subjective, requiring costly software or adapted 2D manual methods that lack reproducibility and analytical rigor. Yet, neurodegenerative disease and aging research demands precise structural comparisons to elucidate mechanisms [...] Read more.

Structural analysis of motoneuron somas and their associated proteins via immunohistochemistry (IHC) remains tedious and subjective, requiring costly software or adapted 2D manual methods that lack reproducibility and analytical rigor. Yet, neurodegenerative disease and aging research demands precise structural comparisons to elucidate mechanisms driving neuronal degeneration. To address this need, we developed a novel algorithm that automates repetitive and subjective IHC analysis tasks, enabling thorough, objective, blinded, order-agnostic, and reproducible 3D batch analysis. With no manual tracing, the algorithm produces 3D Cartesian reconstructions of motoneuron somas from 60× IHC images of mouse lumbar spinal tissue. From these reconstructions, it measures 3D soma volume and efficiently quantitates net somatic protein expression and macro-cluster size. In this validation study, we applied the algorithm to assess soma size and C-bouton expression in various healthy control mice, comparing its measurements against manual measurements and across multiple algorithm users to confirm its accuracy and reproducibility. This novel, customizable tool enables efficient and high-fidelity 3D motoneuron analysis, replacing tedious, qualitative, cell-by-cell manual tuning with automatic threshold adaptation and quantified batch settings. For the first time, we attain reproducible results with quantifiable accuracy, exhaustive sampling, and a high degree of objectivity. Full article

(This article belongs to the Special Issue Data Modeling and Algorithms in Biomedical Applications)

► Show Figures

Figure 1

20 pages, 3966 KiB

Open AccessReview

Mechanotransduction: A Master Regulator of Alveolar Cell Fate Determination

by Kusum Devi and Kalpaj R. Parekh

Bioengineering 2025, 12(7), 760; https://doi.org/10.3390/bioengineering12070760 (registering DOI) - 14 Jul 2025

Abstract

Mechanotransduction plays an essential role in the fate determination of alveolar cells within the pulmonary system by translating mechanical forces into intricate biochemical signals. This process exclusively governs differentiation, phenotypic stability, and maintenance of alveolar epithelial cell subtypes, primarily the alveolar AT1/AT2 cells. [...] Read more.

Mechanotransduction plays an essential role in the fate determination of alveolar cells within the pulmonary system by translating mechanical forces into intricate biochemical signals. This process exclusively governs differentiation, phenotypic stability, and maintenance of alveolar epithelial cell subtypes, primarily the alveolar AT1/AT2 cells. Perturbed mechanical tension proportionally impacts alveolar cell phenotypic identity and their functional characteristics. The fundamental influence of respiratory mechanics on alveolar cell lineage commitment and sustenance is undeniable. AT1 cells are recognized as principal mechanosensors within the alveolus, directly perceiving and responding to mechanical forces imposed by respiration through cell–matrix interactions. These mechanical forces instigate a profound reorganization of the actin cytoskeleton within cells, indispensable for signal transduction and perpetuation of their differentiated phenotype, orchestrated by integrins and cell adhesion molecule-mediated signaling. The dysregulated mechanotransduction in the pulmonary system intrinsically contributes to the etiology and progression of various diseases, exemplified by pulmonary fibrosis. This review systematically elucidates the profound impact of mechanotransduction on alveolar cell differentiation and fate sustenance and underscores how its dysregulation contributes to the initiation and perpetuation of lung diseases. Full article

(This article belongs to the Section Cellular and Molecular Bioengineering)

► Show Figures

Figure 1

22 pages, 3129 KiB

Open AccessArticle

Characterizing the Impact of Fabrication Methods on Mechanically Tunable Gelatin Hydrogels for Cardiac Fibrosis Studies

by Jordyn Folh, Phan Linh Dan Tran and Renita E. Horton

Bioengineering 2025, 12(7), 759; https://doi.org/10.3390/bioengineering12070759 (registering DOI) - 13 Jul 2025

Abstract

The mechanical properties of the extracellular matrix critically influence cell behavior in both physiological and pathophysiological states, including cardiac fibrosis. In vitro models have played a critical role in assessing biological mechanisms. In this study, we characterized mechanically tunable enzymatically crosslinked gelatin-microbial transglutaminase [...] Read more.

The mechanical properties of the extracellular matrix critically influence cell behavior in both physiological and pathophysiological states, including cardiac fibrosis. In vitro models have played a critical role in assessing biological mechanisms. In this study, we characterized mechanically tunable enzymatically crosslinked gelatin-microbial transglutaminase (mTG) hydrogels for modeling cardiovascular diseases. Gelatin hydrogels were fabricated via direct mixing or immersion crosslinking methods. Hydrogel formulations were assessed using the Piuma nanoindenter and Instron systems. This study investigates the effects of fabrication methods, UV ozone (UVO) sterilization, crosslinking methods, and incubation media on hydrogel stiffness. Further, this study examined the response of murine cardiac fibroblasts to hydrogel stiffness. The hydrogels exhibited modulus ranges relevant to both healthy and fibrotic cardiac tissues. UVO exposure led to slight decreases in hydrogel modulus, while the fabrication method had a significant impact on the modulus. Hydrogels incubated in phosphate buffered saline (PBS) were stiffer than those incubated in Medium 199 (M199), which correlated with lower pH in PBS. Fibroblasts cultured on stiffer hydrogels display enhanced smooth muscle actin (SMA) expression, suggesting sensitivity to material stiffness. These findings highlight how fabrication parameters influence the modulus of gelatin-mTG hydrogels for cardiac tissue models. Full article

(This article belongs to the Special Issue Applications of Cell and Tissue Mechanics in Cardiovascular Regenerative)

► Show Figures

Graphical abstract

22 pages, 6526 KiB

Open AccessArticle

Creating Blood Analogs to Mimic Steady-State Non-Newtonian Shear-Thinning Characteristics Under Various Thermal Conditions

by Hang Yi, Alexander Wang, Christopher Wang, Jared Chong, Chungyiu Ma, Luke Bramlage, Bryan Ludwig and Zifeng Yang

Bioengineering 2025, 12(7), 758; https://doi.org/10.3390/bioengineering12070758 (registering DOI) - 12 Jul 2025

Abstract

Blood analogs are widely employed in in vitro experiments such as particle image velocity (PIV) to secure hemodynamics, assisting pathophysiological diagnoses of neurovascular and cardiovascular diseases, as well as pre-surgical planning and intraoperative orientation. To obtain accurate physical parameters, which are critical for [...] Read more.

Blood analogs are widely employed in in vitro experiments such as particle image velocity (PIV) to secure hemodynamics, assisting pathophysiological diagnoses of neurovascular and cardiovascular diseases, as well as pre-surgical planning and intraoperative orientation. To obtain accurate physical parameters, which are critical for diagnosis and treatment, blood analogs should exhibit realistic non-Newtonian shear-thinning features. In this study, two types of blood analogs working under room temperature (293.15 K) were created to mimic the steady-state shear-thinning features of blood over a temperature range of 295 to 312 K and a shear range of 1~250 s⁻¹ at a hematocrit of ~40%. Type I was a general-purpose analog composed of deionized (DI) water and xanthan gum (XG) powder, while Type II was specially designed for PIV tests, incorporating DI water, XG, and fluorescent microspheres. By minimizing the root mean square deviation between generated blood analogs and an established viscosity model, formulas for both blood analogs were successfully derived for the designated temperatures. The results showed that both blood analogs could replicate the shear-thinning viscosities of real blood, with the averaged relative discrepancy < 5%. Additionally, a strong linear correlation was observed between body temperature and XG concentration in both blood analogs (coefficient of determination > 0.96): for Type I, 295–312 K correlates with 140–520 ppm, and for Type II, 295–315 K correlates with 200–560 ppm. This work bridges the gap between idealized steady-state non-Newtonian viscosity models of blood and the complexities of real-world physiological conditions, offering a versatile platform for advancing particle image velocimetry tests and hemodynamics modeling, optimizing therapeutic interventions, and enhancing biomedical technologies in temperature-sensitive environments. Full article

(This article belongs to the Special Issue Pathology–Therapeutics–Medicine of Vascular and Respiratory Diseases: Recent Progress Using CFD and AI)

► Show Figures

Figure 1

12 pages, 3424 KiB

Open AccessArticle

Tri-Layered Full-Thickness Artificial Skin Incorporating Adipose-Derived Stromal Vascular Fraction Cells, Keratinocytes, and a Basement Membrane

by Jung Huh, Seong-Ho Jeong, Eun-Sang Dhong, Seung-Kyu Han and Kyung-Chul Moon

Bioengineering 2025, 12(7), 757; https://doi.org/10.3390/bioengineering12070757 (registering DOI) - 12 Jul 2025

Abstract

Tissue-engineered artificial skin has the potential to enhance wound healing without necessitating extensive surgical procedures or causing donor-site morbidity. The purpose of this study was to examine the possibility of developing tri-layered tissue-engineered full-thickness artificial skin with a basement membrane for clinical use [...] Read more.

Tissue-engineered artificial skin has the potential to enhance wound healing without necessitating extensive surgical procedures or causing donor-site morbidity. The purpose of this study was to examine the possibility of developing tri-layered tissue-engineered full-thickness artificial skin with a basement membrane for clinical use to accelerate wound healing. We engineered full-thickness artificial skin with a basement membrane for wound healing by employing stromal vascular fraction (SVF) cells for the dermal layer and autologous keratinocytes for the epidermal layer. The fabrication of a basement membrane involved the use of 100% bovine collagen and 4% elastin produced through a low-temperature three-dimensional printer. Scaffolds for cells were printed with 100% bovine collagen. The basement membrane underwent evaluations for collagenase degradation, tensile strength, and structural characteristics using scanning electron microscopy. The final tri-layered full-thickness artificial skin included two cell scaffolds with a basement membrane between them. The basement membrane may support cellular attachment without inducing significant cytotoxic effects. This study presents a novel strategy for full-thickness artificial skin development, combining SVF and keratinocytes with an optimized collagen-elastin basement membrane. This method may overcome the significant limitations of current artificial skin, thereby contributing to the advancement of tissue-engineering in wound healing for clinical use. Full article

(This article belongs to the Section Biomedical Engineering and Biomaterials)

► Show Figures

Figure 1

18 pages, 24638 KiB

Open AccessArticle

Accelerating Wound Healing Through Deep Reinforcement Learning: A Data-Driven Approach to Optimal Treatment

by Fan Lu, Ksenia Zlobina, Prabhat Baniya, Houpu Li, Nicholas Rondoni, Narges Asefifeyzabadi, Wan Shen Hee, Maryam Tebyani, Kaelan Schorger, Celeste Franco, Michelle Bagood, Mircea Teodorescu, Marco Rolandi, Rivkah Isseroff and Marcella Gomez

Bioengineering 2025, 12(7), 756; https://doi.org/10.3390/bioengineering12070756 (registering DOI) - 11 Jul 2025

Abstract

Advancements in bioelectronic sensors and actuators have paved the way for real-time monitoring and control of the progression of wound healing. Real-time monitoring allows for precise adjustment of treatment strategies to align them with an individual’s unique biological response. However, due to the [...] Read more.

Advancements in bioelectronic sensors and actuators have paved the way for real-time monitoring and control of the progression of wound healing. Real-time monitoring allows for precise adjustment of treatment strategies to align them with an individual’s unique biological response. However, due to the complexities of human–drug interactions and a lack of predictive models, it is challenging to determine how one should adjust drug dosage to achieve the desired biological response. This work proposes an adaptive closed-loop control framework that integrates deep learning, optimal control, and reinforcement learning to update treatment strategies in real time, with the goal of accelerating wound closure. The proposed approach eliminates the need for mathematical modeling of complex nonlinear wound-healing dynamics. We demonstrate the convergence of the controller via an in silico experimental setup, where the proposed approach successfully accelerated the wound-healing process by 17.71%. Finally, we share the experimental setup and results of an in vivo implementation to highlight the translational potential of our work. Our data-driven model suggests that the treatment strategy, as determined by our deep reinforcement learning algorithm, results in an accelerated onset of inflammation and subsequent transition to proliferation in a porcine wound model. Full article

(This article belongs to the Special Issue Machine Learning and Artificial Intelligence for Biomedical Applications, 3rd Edition)

► Show Figures

Figure 1

18 pages, 1442 KiB

Open AccessArticle

Hyperspectral Imaging for Enhanced Skin Cancer Classification Using Machine Learning

by Teng-Li Lin, Arvind Mukundan, Riya Karmakar, Praveen Avala, Wen-Yen Chang and Hsiang-Chen Wang

Bioengineering 2025, 12(7), 755; https://doi.org/10.3390/bioengineering12070755 (registering DOI) - 11 Jul 2025

Abstract

Objective: The classification of skin cancer is very helpful in its early diagnosis and treatment, considering the complexity involved in differentiating AK from BCC and SK. These conditions are generally not easily detectable due to their comparable clinical presentations. Method: This paper presents [...] Read more.

Objective: The classification of skin cancer is very helpful in its early diagnosis and treatment, considering the complexity involved in differentiating AK from BCC and SK. These conditions are generally not easily detectable due to their comparable clinical presentations. Method: This paper presents a new approach to hyperspectral imaging for enhancing the visualization of skin lesions called the Spectrum-Aided Vision Enhancer (SAVE), which has the ability to convert any RGB image into a narrow-band image (NBI) by combining hyperspectral imaging (HSI) to increase the contrast of the area of the cancerous lesions when compared with the normal tissue, thereby increasing the accuracy of classification. The current study investigates the use of ten different machine learning algorithms for the purpose of classification of AK, BCC, and SK, including convolutional neural network (CNN), random forest (RF), you only look once (YOLO) version 8, support vector machine (SVM), ResNet50, MobileNetV2, Logistic Regression, SVM with stochastic gradient descent (SGD) Classifier, SVM with logarithmic (LOG) Classifier and SVM- Polynomial Classifier, in assessing the capability of the system to differentiate AK from BCC and SK with heightened accuracy. Results: The results demonstrated that SAVE enhanced classification performance and increased its accuracy, sensitivity, and specificity compared to a traditional RGB imaging approach. Conclusions: This advanced method offers dermatologists a tool for early and accurate diagnosis, reducing the likelihood of misclassification and improving patient outcomes. Full article

(This article belongs to the Special Issue Biomedical Imaging and Data Analytics for Disease Diagnosis and Treatment, 3rd Edition)

► Show Figures

Figure 1

10 pages, 206 KiB

Open AccessArticle

AI-Enhanced 3D Transperineal Ultrasound: Advancing Biometric Measurements for Precise Prolapse Severity Assessment

by Desirèe De Vicari, Marta Barba, Alice Cola, Clarissa Costa, Mariachiara Palucci and Matteo Frigerio

Bioengineering 2025, 12(7), 754; https://doi.org/10.3390/bioengineering12070754 - 11 Jul 2025

Abstract

Pelvic organ prolapse (POP) is a common pelvic floor disorder with substantial impact on women’s quality of life, necessitating accurate and reproducible diagnostic methods. This study investigates the use of three-dimensional (3D) transperineal ultrasound, integrated with artificial intelligence (AI), to evaluate pelvic floor [...] Read more.

Pelvic organ prolapse (POP) is a common pelvic floor disorder with substantial impact on women’s quality of life, necessitating accurate and reproducible diagnostic methods. This study investigates the use of three-dimensional (3D) transperineal ultrasound, integrated with artificial intelligence (AI), to evaluate pelvic floor biomechanics and identify correlations between biometric parameters and prolapse severity. Thirty-seven female patients diagnosed with genital prolapse (mean age: 65.3 ± 10.6 years; mean BMI: 29.5 ± 3.8) were enrolled. All participants underwent standardized 3D transperineal ultrasound using the Mindray Smart Pelvic system, an AI-assisted imaging platform. Key biometric parameters—anteroposterior diameter, laterolateral diameter, and genital hiatus area—were measured under three functional states: rest, maximal Valsalva maneuver, and voluntary pelvic floor contraction. Additionally, two functional indices were derived: the distensibility index (ratio of Valsalva to rest) and the contractility index (ratio of contraction to rest), reflecting pelvic floor elasticity and muscular function, respectively. Statistical analysis included descriptive statistics and univariate correlation analysis using Pelvic Organ Prolapse Quantification (POP-Q) system scores. Results revealed a significant correlation between laterolateral diameter and prolapse severity across multiple compartments and functional states. In apical prolapse, the laterolateral diameter measured at rest and during both Valsalva and contraction showed positive correlations with POP-Q point C, indicating increasing transverse pelvic dimensions with more advanced prolapse (e.g., r = 0.42 to 0.58; p < 0.05). In anterior compartment prolapse, the same parameter measured during Valsalva and contraction correlated significantly with POP-Q point AA (e.g., r = 0.45 to 0.61; p < 0.05). Anteroposterior diameters and genital hiatus area were also analyzed but showed weaker or inconsistent correlations. AI integration facilitated real-time image segmentation and automated measurement, reducing operator dependency and increasing reproducibility. These findings highlight the laterolateral diameter as a strong, reproducible anatomical marker for POP severity, particularly when assessed dynamically. The combined use of AI-enhanced imaging and functional indices provides a novel, standardized, and objective approach for assessing pelvic floor dysfunction. This methodology supports more accurate diagnosis, individualized management planning, and long-term monitoring of pelvic floor disorders. Full article

(This article belongs to the Special Issue Artificial Intelligence and Wearable Biosensors for Precision Medicine: Integrating Technology)

5 pages, 144 KiB

Open AccessEditorial

Special Issue: “Artificial Intelligence for Biomedical Signal Processing”

by Fernando Vaquerizo-Villar and Verónica Barroso-García

Bioengineering 2025, 12(7), 753; https://doi.org/10.3390/bioengineering12070753 - 10 Jul 2025

Abstract

Artificial intelligence (AI) has rapidly evolved from a conceptual promise into a practical and transformative asset in modern medicine [...] Full article

(This article belongs to the Special Issue Artificial Intelligence for Biomedical Signal Processing)

22 pages, 3299 KiB

Open AccessArticle

Lokomat-Assisted Robotic Rehabilitation in Spinal Cord Injury: A Biomechanical and Machine Learning Evaluation of Functional Symmetry and Predictive Factors

by Alexandru Bogdan Ilies, Cornel Cheregi, Hassan Hassan Thowayeb, Jan Reinald Wendt, Maur Sebastian Horgos and Liviu Lazar

Bioengineering 2025, 12(7), 752; https://doi.org/10.3390/bioengineering12070752 - 10 Jul 2025

Abstract

Background: Lokomat-assisted robotic rehabilitation is increasingly used for gait restoration in patients with spinal cord injury (SCI). However, the objective evaluation of treatment effectiveness through biomechanical parameters and machine learning approaches remains underexplored. Methods: This study analyzed data from 29 SCI patients undergoing [...] Read more.

Background: Lokomat-assisted robotic rehabilitation is increasingly used for gait restoration in patients with spinal cord injury (SCI). However, the objective evaluation of treatment effectiveness through biomechanical parameters and machine learning approaches remains underexplored. Methods: This study analyzed data from 29 SCI patients undergoing Lokomat-based rehabilitation. A dataset of 46 variables including range of motion (L-ROM), joint stiffness (L-STIFF), and muscular force (L-FORCE) was examined using statistical methods (paired t-test, ANOVA, and ordinary least squares regression), clustering techniques (k-means), dimensionality reduction (t-SNE), and anomaly detection (Isolation Forest). Predictive modeling was applied to assess the influence of age, speed, body weight, body weight support, and exercise duration on biomechanical outcomes. Results: No statistically significant asymmetries were found between left and right limb measurements, indicating functional symmetry post-treatment (p > 0.05). Clustering analysis revealed a weak structure among patient groups (Silhouette score ≈ 0.31). Isolation Forest identified minimal anomalies in stiffness data, supporting treatment consistency. Regression models showed that body weight and body weight support significantly influenced joint stiffness (p < 0.01), explaining up to 60% of the variance in outcomes. Conclusions: Lokomat-assisted robotic rehabilitation demonstrates high functional symmetry and biomechanical consistency in SCI patients. Machine learning methods provided meaningful insight into the structure and predictability of outcomes, highlighting the clinical value of weight and support parameters in tailoring recovery protocols. Full article

(This article belongs to the Special Issue Regenerative Rehabilitation for Spinal Cord Injury)

► Show Figures

Figure 1

24 pages, 1680 KiB

Open AccessArticle

Mathematical Modeling of Salmonella Cancer Therapies Demonstrates the Necessity of Both Bacterial Cytotoxicity and Immune Activation

by Lars M. Howell and Neil S. Forbes

Bioengineering 2025, 12(7), 751; https://doi.org/10.3390/bioengineering12070751 - 10 Jul 2025

Abstract

Salmonella therapies are a promising tool for the treatment of solid tumors. Salmonella can be engineered to increase their tumor infiltration, cell killing abilities, and immunostimulatory properties. However, bacterial therapies have often failed in clinical trials due to poor characterization. Mathematical models are [...] Read more.

Salmonella therapies are a promising tool for the treatment of solid tumors. Salmonella can be engineered to increase their tumor infiltration, cell killing abilities, and immunostimulatory properties. However, bacterial therapies have often failed in clinical trials due to poor characterization. Mathematical models are useful for predicting the immune response to cancer treatments and characterizing the properties of bacterial invasion. Herein we develop an ordinary differential equation-based model that combines bacterial therapies with classical anti-tumor immunotherapies. Our modeling results suggest that increasing bacterial localization to the tumor is key for therapeutic efficacy; however, increased intracellular invasion and direct bacterial mediated cytotoxicity does not reduce tumor growth. Further, the model suggests that enhancing T cell-mediated cell death by both bacterial stimulation of pro-inflammatory cytokines and activation of T cells via antigen cascade is critical for therapeutic efficacy. A balance of intracellular and extracellular Salmonella leads to more effective therapeutic response, which suggests a strategy for strain design to be tested in vivo. Overall, this model provides a system to predict which engineered features of Salmonella therapies lead to effective treatment outcomes. Full article

(This article belongs to the Special Issue The Power of Models and Simulation Tools in Biomedical and Biochemical Engineering)

► Show Figures

Figure 1

19 pages, 259 KiB

Open AccessArticle

Understanding the Impact of Assistive Technology on Users’ Lives in England: A Capability Approach

by Rebecca Joskow, Dilisha Patel, Anna Landre, Kate Mattick, Catherine Holloway, Jamie Danemayer and Victoria Austin

Bioengineering 2025, 12(7), 750; https://doi.org/10.3390/bioengineering12070750 - 9 Jul 2025

Abstract

This study presents an analysis of England’s 2023 national assessment of assistive technology (AT) access and use, with a particular focus on the qualitative impact of AT as described by users. It aims to address limitations in conventional AT impact assessments, which often [...] Read more.

This study presents an analysis of England’s 2023 national assessment of assistive technology (AT) access and use, with a particular focus on the qualitative impact of AT as described by users. It aims to address limitations in conventional AT impact assessments, which often prioritize clinical outcomes or user satisfaction, by offering a deeper account of how impact is experienced in everyday life. Drawing on data from a nationally representative survey of 7000 disabled adults and children, as well as six focus group discussions and 28 semi-structured interviews with stakeholders across the WHO 5Ps framework (People, Providers, Personnel, Policy, and Products), the study applies Amartya Sen and Martha Nussbaum’s Capability Approach to explore these experiences. Using inductive thematic analysis, we identify three main domains of user-reported impact: Functions and Activities (e.g., mobility, communication, vision, leisure, daily routines, and cognitive support), Outcomes (e.g., autonomy, quality of life, safety, social participation, wellbeing, and work and learning), and Lived Experience (e.g., access barriers, essentiality, identity and emotional connection, peace of mind, and sense of control and confidence). These findings offer a more user-centered understanding of AT impact and can inform the development of future measurement tools, research design, and government-led interventions to improve AT provision. Full article

(This article belongs to the Special Issue Measuring Outcomes and Impact Related to Assistive Technology and Accessibility for Disability)

18 pages, 1196 KiB

Open AccessArticle

Machine Learning Approaches for Early Detection of Ossification of Posterior Longitudinal Ligament in Health Screening Settings

by Ryo Mizukoshi, Ryosuke Maruiwa, Keitaro Ito, Norihiro Isogai, Haruki Funao, Retsu Fujita and Mitsuru Yagi

Bioengineering 2025, 12(7), 749; https://doi.org/10.3390/bioengineering12070749 - 9 Jul 2025

Abstract

Early detection of ossification of the posterior longitudinal ligament (OPLL) is hampered by the late onset of neurological symptoms, so we built and validated an interpretable machine learning model to identify OPLL during routine health examinations. We retrospectively analyzed 1442 Japanese adults screened [...] Read more.

Early detection of ossification of the posterior longitudinal ligament (OPLL) is hampered by the late onset of neurological symptoms, so we built and validated an interpretable machine learning model to identify OPLL during routine health examinations. We retrospectively analyzed 1442 Japanese adults screened between 2020 and 2023, including 432 imaging-confirmed cases, after median imputation, one-hot encoding, Random Forest feature selection that reduced 235 variables to 20, and class-balance correction with SMOTE. Logistic regression, Random Forest, Gradient Boosting, and XGBoost models were tuned using a 5-fold cross-validated grid search, in which a re-estimated logistic regression yielded odds ratios for clinical interpretation. The logistic model achieved 65% accuracy and an AUROC of 0.69 (95% CI 0.66–0.76), matching tree-based models, yet with fewer false-negatives. Advanced age (OR 1.60, 95% CI 1.27–2.00) and elevated CA19-9 (OR 1.24, 95% CI 1.00–1.35) independently increased OPLL odds. This concise, explainable tool could facilitate early recognition of OPLL, reduce unnecessary follow-up, and enable timely preventive interventions in high-volume screening programs. Full article

(This article belongs to the Section Biomedical Engineering and Biomaterials)

► Show Figures

Figure 1

17 pages, 3961 KiB

Open AccessArticle

Therapeutic Potential of Local Application of Fibroblast Growth Factor-2 to Periodontal Defects in a Preclinical Osteoporosis Model

by Shinta Mori, Sho Mano, Naoki Miyata, Tasuku Murakami, Wataru Yoshida, Kentaro Imamura and Atsushi Saito

Bioengineering 2025, 12(7), 748; https://doi.org/10.3390/bioengineering12070748 - 9 Jul 2025

Abstract

This study investigated the effects of local fibroblast growth factor (FGF)-2 application on periodontal healing in an osteoporotic model, both in vivo and in vitro. Wistar rats were divided into the ovariectomy (OVX) and Control groups. Periodontal defects were created 8 weeks post-OVX [...] Read more.

This study investigated the effects of local fibroblast growth factor (FGF)-2 application on periodontal healing in an osteoporotic model, both in vivo and in vitro. Wistar rats were divided into the ovariectomy (OVX) and Control groups. Periodontal defects were created 8 weeks post-OVX and treated with hydroxypropylcellulose (HPC) or FGF-2 + HPC. Healing was evaluated through micro-computed tomography and histological analyses at 2 and 4 weeks. In vitro, bone marrow mesenchymal stromal cells (BMSCs) were cultured with/without FGF-2 and assessed for cell morphology, viability/proliferation, and osteoblastic marker expression. Alkaline phosphatase (ALP) staining was also performed. FGF-2-treated defects in both groups showed significantly greater bone volume fraction, trabecular number, and thickness compared to HPC only. Histologically, FGF-2 enhanced new bone formation, with the greatest levels in the Control group. In vitro, OVX BMSCs showed reduced actin staining versus controls. FGF-2 increased cell viability/proliferation and protrusions in both groups while downregulating Alpl and Bglap expression levels and reducing ALP-positive cells. FGF-2 increased new bone formation in the OVX group, stimulated proliferation of OVX BMSCs, and modulated their differentiation. FGF-2 could enhance periodontal healing even under osteoporotic conditions, albeit to a lesser extent. Full article

(This article belongs to the Special Issue Recent Advances in Periodontal Tissue Engineering)

► Show Figures

Graphical abstract

30 pages, 4254 KiB

Open AccessArticle

Hybrid Attention-Enhanced Xception and Dynamic Chaotic Whale Optimization for Brain Tumor Diagnosis

by Aliyu Tetengi Ibrahim, Ibrahim Hayatu Hassan, Mohammed Abdullahi, Armand Florentin Donfack Kana, Amina Hassan Abubakar, Mohammed Tukur Mohammed, Lubna A. Gabralla, Mohamad Khoiru Rusydi and Haruna Chiroma

Bioengineering 2025, 12(7), 747; https://doi.org/10.3390/bioengineering12070747 - 9 Jul 2025

Abstract

In medical diagnostics, brain tumor classification remains essential, as accurate and efficient models aid medical professionals in early detection and treatment planning. Deep learning methodologies for brain tumor classification have gained popularity due to their potential to deliver prompt and precise diagnostic results. [...] Read more.

In medical diagnostics, brain tumor classification remains essential, as accurate and efficient models aid medical professionals in early detection and treatment planning. Deep learning methodologies for brain tumor classification have gained popularity due to their potential to deliver prompt and precise diagnostic results. This article proposes a novel classification technique that integrates the Xception model with a hybrid attention mechanism and progressive image resizing to enhance performance. The methodology is built on a combination of preprocessing techniques, transfer learning architecture reconstruction, and dynamic fine-tuning strategies. To optimize key hyper-parameters, this study employed the Dynamic Chaotic Whale Optimization Algorithm. Additionally, we developed a novel learning rate scheduler that dynamically adjusts the learning rate based on image size at each training phase, improving training efficiency and model adaptability. Batch sizes and layer freezing methods were also adjusted according to image size. We constructed an ensemble approach by preserving models trained on different image sizes and merging their results using weighted averaging, bagging, boosting, stacking, blending, and voting techniques. Our proposed method was evaluated on benchmark datasets achieving remarkable accuracies of 99.67%, 99.09%, and 99.67% compared to the classical algorithms. Full article

(This article belongs to the Special Issue Recent Advances in Machine Learning and Explainable Artificial Intelligence in Biomedical Data Mining, and Disease Diagnosis Frameworks)

► Show Figures

Figure 1

11 pages, 3262 KiB

Open AccessArticle

Evaluation of Mandibular Bone Alterations by Panoramic Radiography: A Potential Tool in the Identification of Signs of Osteopenia and Osteoporosis

by Esdras Gabriel Alves-Silva, Betania Fachetti Ribeiro, Camila Fontes Silva, Rita de Kássia-Alves, Rodrigo Arruda-Vasconcelos, Lidiane Mendes Louzada, Rebecca F. Almeida-Gomes, João Miguel Marques Santos and Brenda P. F. A. Gomes

Bioengineering 2025, 12(7), 746; https://doi.org/10.3390/bioengineering12070746 - 9 Jul 2025

Abstract

This study aimed to evaluate the validity of panoramic radiography as an auxiliary method for identifying mandibular bone features consistent with a diagnosis of osteopenia or osteoporosis. Ninety panoramic radiographs were analyzed to assess the quality of the mandibular cortical layer below the [...] Read more.

This study aimed to evaluate the validity of panoramic radiography as an auxiliary method for identifying mandibular bone features consistent with a diagnosis of osteopenia or osteoporosis. Ninety panoramic radiographs were analyzed to assess the quality of the mandibular cortical layer below the mental foramen on both sides of the mandible. Scores C1 (normal), C2 (osteopenia), and C3 (osteoporosis) were attributed according to the cortical morphology. The sample consisted of 78 (86%) women aged 45 years or older and 12 (14%) men older than 60 years old. In 39 (43%) cases, the C1 score was evidenced as the lower mandibular cortical layer was normal on the image. The C2 score was identified in 47 (52%) cases, in which the cortical layer showed semilunar defects. Four (5%) cases presented a C3 score, with the cortical layer showing a clearly porous, thinner bone cortex. The presence of risk behaviors (e.g., smoking and alcoholism) as well as some comorbidities (e.g., systemic arterial hypertension, diabetes mellitus and thyroid disorders) was also observed. Mandibular bone changes were observed in association with a set of risk factors using panoramic radiography. Full article

(This article belongs to the Special Issue Bioengineered Approaches to Oral-Systemic Disease Diagnostics and Therapeutics)

► Show Figures

Figure 1

Journal Description

Bioengineering

Latest Articles

Journal Menu

Journal Browser

Highly Accessed Articles

Latest Books

E-Mail Alert

News

Topics

Conferences

Special Issues

Topical Collections

Further Information

Guidelines

MDPI Initiatives

Follow MDPI