Bioengineering

15 pages, 5911 KB

Open AccessArticle

Integrative Bioinformatics-Guided Analysis of Glomerular Transcriptome Implicates Potential Therapeutic Targets and Pathogenesis Mechanisms in IgA Nephropathy

by Tiange Yang, Mengde Dai, Fen Zhang and Weijie Wen

Bioengineering 2025, 12(10), 1040; https://doi.org/10.3390/bioengineering12101040 (registering DOI) - 27 Sep 2025

Abstract

(1) Background: IgA nephropathy (IgAN) is a leading cause of chronic kidney disease worldwide. Despite its prevalence, the molecular mechanisms of IgAN remain poorly understood, partly due to limited research scale. Identifying key genes involved in IgAN’s pathogenesis is critical for novel diagnostic [...] Read more.

(1) Background: IgA nephropathy (IgAN) is a leading cause of chronic kidney disease worldwide. Despite its prevalence, the molecular mechanisms of IgAN remain poorly understood, partly due to limited research scale. Identifying key genes involved in IgAN’s pathogenesis is critical for novel diagnostic and therapeutic strategies. (2) Methods: We identified differentially expressed genes (DEGs) by analyzing public datasets from the Gene Expression Omnibus. Gene Ontology and Kyoto Encyclopedia of Genes and Genomes analyses were performed to elucidate the biological roles of DEGs. Hub genes were screened using weighted gene co-expression network analysis combined with machine learning algorithms. Immune infiltration analysis was conducted to explore associations between hub genes and immune cell profiles. The hub genes were validated using receiver operating characteristic curves and area under the curve. (3) Results: We identified 165 DEGs associated with IgAN and revealed pathways such as IL-17 signaling and complement and coagulation cascades, and biological processes including response to xenobiotic stimuli. Four hub genes were screened: three downregulated (FOSB, SLC19A2, PER1) and one upregulated (SOX17). The AUC values for identifying IgAN in the training and testing set ranged from 0.956 to 0.995. Immune infiltration analysis indicated that hub gene expression correlated with immune cell abundance, suggesting their involvement in IgAN’s immune pathogenesis. (4) Conclusion: This study identifies FOSB, SLC19A2, PER1, and SOX17 as novel hub genes with high diagnostic accuracy for IgAN. These genes, linked to immune-related pathways such as IL-17 signaling and complement activation, offer promising targets for diagnostic development and therapeutic intervention, enhancing our understanding of IgAN’s molecular and immune mechanisms. Full article

(This article belongs to the Special Issue Advanced Biomedical Signal Communication Technology)

► Show Figures

Figure 1

22 pages, 1278 KB

Open AccessReview

Artificial Intelligence-Based Methods and Omics for Mental Illness Diagnosis: A Review

by Glenda Santos de Oliveira, Fábio Henrique dos Santos Rodrigues, João Guilherme de Moraes Pontes and Ljubica Tasic

Bioengineering 2025, 12(10), 1039; https://doi.org/10.3390/bioengineering12101039 (registering DOI) - 27 Sep 2025

Abstract

The underlying causes fof major mental illnesses, including anxiety disorders (ADs), depression, and bipolar disorder (BD), remain insufficiently understood, limiting the availability of effective, patient-friendly treatments and accurate diagnostic tests. For instance, anxiety disorders encompass a diverse spectrum of subtypes and may emerge [...] Read more.

The underlying causes fof major mental illnesses, including anxiety disorders (ADs), depression, and bipolar disorder (BD), remain insufficiently understood, limiting the availability of effective, patient-friendly treatments and accurate diagnostic tests. For instance, anxiety disorders encompass a diverse spectrum of subtypes and may emerge at different stages of mental illness, each with distinct symptom profiles. This heterogeneity often complicates differential diagnosis, leading, in many cases, to delayed treatment or inappropriate management. In recent years, technological advances have enabled the development of artificial intelligence (AI)-based approaches that, when integrated with multi-omics data, offer substantial advantages over traditional statistical methods, particularly for analysing large-scale datasets and integrating clinical with bioanalytical information. This review analyses current efforts to identify biomarkers for mental illness and explores the application of machine learning, deep learning, and computational modelling in advancing personalised and precise diagnostics. Full article

(This article belongs to the Special Issue Smart Engineering: Integrating Artificial Intelligence and Bioengineering)

► Show Figures

Graphical abstract

21 pages, 1493 KB

Open AccessSystematic Review

From Echocardiography to CT/MRI: Lessons for AI Implementation in Cardiovascular Imaging in LMICs—A Systematic Review and Narrative Synthesis

by Ahmed Marey, Saba Mehrtabar, Ahmed Afify, Basudha Pal, Arcadia Trvalik, Sola Adeleke and Muhammad Umair

Bioengineering 2025, 12(10), 1038; https://doi.org/10.3390/bioengineering12101038 (registering DOI) - 27 Sep 2025

Abstract

Objectives: The aim of this study was to synthesize current evidence on artificial intelligence (AI) adoption in cardiovascular imaging across low- and middle-income countries (LMICs), highlighting diagnostic performance, implementation barriers, and potential solutions. Methods: We conducted a systematic review of PubMed, [...] Read more.

Objectives: The aim of this study was to synthesize current evidence on artificial intelligence (AI) adoption in cardiovascular imaging across low- and middle-income countries (LMICs), highlighting diagnostic performance, implementation barriers, and potential solutions. Methods: We conducted a systematic review of PubMed, Embase, Cochrane Library, Web of Science, and Scopus for studies evaluating AI-based echocardiography, cardiac CT, or cardiac MRI in LMICs. Articles were screened according to PRISMA guidelines, and data on diagnostic outcomes, challenges, and enabling factors were extracted and narratively synthesized. Results: Twelve studies met the inclusion criteria. AI-driven methods frequently surpassed 90% accuracy in detecting coronary artery disease, rheumatic heart disease, and left ventricular hypertrophy, often enabling task shifting to non-expert operators. Challenges included limited dataset diversity, operator dependence, infrastructure constraints, and ethical considerations. Insights from high-income countries, such as automated segmentation and accelerated imaging, suggest potential for broader AI integration in cardiac MRI and CT. Conclusions: AI holds promise for enhancing cardiovascular care in LMICs by improving diagnostic accuracy and workforce efficiency. However, multi-center data sharing, targeted training, reliable infrastructure, and robust governance are essential for sustainable adoption. This review underscores AI’s capacity to bridge resource gaps in LMICs, offering practical pathways for future research, clinical practice, and policy development in global cardiovascular imaging. Full article

(This article belongs to the Special Issue Artificial Intelligence Revolution in Biomedical Image and Signal Processing: Innovations and Applications)

22 pages, 996 KB

Open AccessSystematic Review

Integrating Spatial Omics and Deep Learning: Toward Predictive Models of Cardiomyocyte Differentiation Efficiency

by Tumo Kgabeng, Lulu Wang, Harry M. Ngwangwa and Thanyani Pandelani

Bioengineering 2025, 12(10), 1037; https://doi.org/10.3390/bioengineering12101037 (registering DOI) - 27 Sep 2025

Abstract

Advances in cardiac regenerative medicine increasingly rely on integrating artificial intelligence with spatial multi-omics technologies to decipher intricate cellular dynamics in cardiomyocyte differentiation. This systematic review, synthetising insights from 88 PRISMA selected studies spanning 2015–2025, explores how deep learning architectures, specifically Graph Neural [...] Read more.

Advances in cardiac regenerative medicine increasingly rely on integrating artificial intelligence with spatial multi-omics technologies to decipher intricate cellular dynamics in cardiomyocyte differentiation. This systematic review, synthetising insights from 88 PRISMA selected studies spanning 2015–2025, explores how deep learning architectures, specifically Graph Neural Networks (GNNs) and Recurrent Neural Networks (RNNs), synergise with multi-modal single-cell datasets, spatially resolved transcriptomics, and epigenomics to advance cardiac biology. Innovations in spatial omics technologies have revolutionised our understanding of the organisation of cardiac tissue, revealing novel cellular communities and metabolic landscapes that underlie cardiovascular health and disease. By synthesising cutting-edge methodologies and technical innovations across these 88 studies, this review establishes the foundation for AI-enabled cardiac regeneration, potentially accelerating the clinical adoption of regenerative treatments through improved therapeutic prediction models and mechanistic understanding. We examine deep learning implementations in spatiotemporal genomics, spatial multi-omics applications in cardiac tissues, cardiomyocyte differentiation challenges, and predictive modelling innovations that collectively advance precision cardiology and next-generation regenerative strategies. Full article

(This article belongs to the Special Issue Artificial Intelligence for Computer-Aided Detection in Biomedical Applications)

17 pages, 5124 KB

Open AccessArticle

Self-Attention Diffusion Models for Zero-Shot Biomedical Image Segmentation: Unlocking New Frontiers in Medical Imaging

by Abderrachid Hamrani and Anuradha Godavarty

Bioengineering 2025, 12(10), 1036; https://doi.org/10.3390/bioengineering12101036 (registering DOI) - 27 Sep 2025

Abstract

Producing high-quality segmentation masks for medical images is a fundamental challenge in biomedical image analysis. Recent research has investigated the use of supervised learning with large volumes of labeled data to improve segmentation across medical imaging modalities and unsupervised learning with unlabeled data [...] Read more.

Producing high-quality segmentation masks for medical images is a fundamental challenge in biomedical image analysis. Recent research has investigated the use of supervised learning with large volumes of labeled data to improve segmentation across medical imaging modalities and unsupervised learning with unlabeled data to segment without detailed annotations. However, a significant hurdle remains in constructing a model that can segment diverse medical images in a zero-shot manner without any annotations. In this work, we introduce the attention diffusion zero-shot unsupervised system (ADZUS), a new method that uses self-attention diffusion models to segment biomedical images without needing any prior labels. This method combines self-attention mechanisms to enable context-aware and detail-sensitive segmentations, with the strengths of the pre-trained diffusion model. The experimental results show that ADZUS outperformed state-of-the-art models on various medical imaging datasets, such as skin lesions, chest X-ray infections, and white blood cell segmentations. The model demonstrated significant improvements by achieving Dice scores ranging from 88.7% to 92.9% and IoU scores from 66.3% to 93.3%. The success of the ADZUS model in zero-shot settings could lower the costs of labeling data and help it adapt to new medical imaging tasks, improving the diagnostic capabilities of AI-based medical imaging technologies. Full article

(This article belongs to the Special Issue Medical Imaging Analysis: Current and Future Trends)

► Show Figures

Figure 1

19 pages, 2814 KB

Open AccessArticle

Verification of the Effectiveness of a Token Economy Method Through Digital Intervention Content for Children with Attention-Deficit/Hyperactivity Disorder

by Seon-Chil Kim

Bioengineering 2025, 12(10), 1035; https://doi.org/10.3390/bioengineering12101035 (registering DOI) - 26 Sep 2025

Abstract

Recently, cognitive training programs using digital content with visuoperceptual stimulation have been developed and commercialized. In particular, digital intervention content for children with attention deficit hyperactivity disorder (ADHD) has been developed as games, enhancing motivation and accessibility for the target population. Active stimulation [...] Read more.

Recently, cognitive training programs using digital content with visuoperceptual stimulation have been developed and commercialized. In particular, digital intervention content for children with attention deficit hyperactivity disorder (ADHD) has been developed as games, enhancing motivation and accessibility for the target population. Active stimulation is required to elicit positive effects on self-regulation training, including attention control and impulse inhibition, through task-based content. Common forms of stimulation include emotional stimuli, such as praise and encouragement, and economic stimuli based on a self-directed token economy system. Economic stimulation can serve as active reinforcement because the child directly engages as the primary agent within the task content. This study applied and validated a token economy intervention using digital therapeutic content in children with ADHD. Behavioral assessments were conducted using the Comprehensive Attention Test (CAT) and the Korean version of the Child Behavior Checklist (K-CBCL). The developed digital intervention content implemented a user-centered token economy based on points within the program. In the CAT Flanker Task, the experimental group (0.84 ± 0.40) showed significantly higher sensitivity factor scores than the control group (0.72 ± 0.59) after 4 weeks, with a large effect size (F = 4.76, p = 0.038, partial η² = 0.150). Additionally, the rate of change in externalizing behavior scores on the K-CBCL showed a significant difference between the two groups (t = 2.35, p = 0.026, Cohen’s d = 0.860), demonstrating greater improvement in externalizing symptoms in the experimental group than in the control group. Therefore, this study suggests that the participant-centered implementation model using token economy mechanisms in digital intervention content may serve as a novel and effective therapeutic approach for children with ADHD. Full article

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

12 pages, 538 KB

Open AccessArticle

Gait and Postural Control Deficits in Diabetic Patients with Peripheral Neuropathy Compared to Healthy Controls

by Safi Ullah, Kamran Iqbal and Muhammad Rizwan

Bioengineering 2025, 12(10), 1034; https://doi.org/10.3390/bioengineering12101034 (registering DOI) - 26 Sep 2025

Abstract

Diabetic peripheral neuropathy (DPN) is a common complication of type 2 diabetes that impairs gait and balance, increasing fall risk. This study investigated gait characteristics and postural control in individuals with DPN, compared to age- and gender-matched healthy controls. Fifteen DPN patients and [...] Read more.

Diabetic peripheral neuropathy (DPN) is a common complication of type 2 diabetes that impairs gait and balance, increasing fall risk. This study investigated gait characteristics and postural control in individuals with DPN, compared to age- and gender-matched healthy controls. Fifteen DPN patients and fifteen controls underwent assessments of gait, static balance, and mobility. Gait parameters were measured during overground walking using motion capture and force platforms. Static balance was evaluated via tandem stance tests (eyes open/closed), while mobility was assessed with the Timed-Up-and-Go (TUG) test. Dynamic stability was assessed by computing the center-of-pressure Time-to-Contact (TTC) with the mediolateral (ML) stability boundary. We hypothesized that patients with DPN would exhibit an altered gait and reduced ML postural stability during walking. The study results show no significant differences in ML center-of-pressure (COP) excursion or its velocity during walking between groups. Patients with DPN walked relatively slowly, with shorter steps, and showed markedly poorer static balance (earlier failure during tandem stance test), as well as slower TUG performance. Clinically, these findings support routine fall risk screening in DPN using both static balance tests (e.g., tandem stance) and mobility measures (e.g., TUG or gait speed). These findings further suggest that while dynamic postural control during walking may be preserved, DPN patients exhibit gait adaptations and significant static balance deficits, highlighting the need for comprehensive balance assessment in this population. Full article

(This article belongs to the Special Issue Biomechanics in Sport and Motion Analysis)

► Show Figures

Figure 1

20 pages, 1488 KB

Open AccessArticle

Attention-Fusion-Based Two-Stream Vision Transformer for Heart Sound Classification

by Kalpeshkumar Ranipa, Wei-Ping Zhu and M. N. S. Swamy

Bioengineering 2025, 12(10), 1033; https://doi.org/10.3390/bioengineering12101033 - 26 Sep 2025

Abstract

Vision Transformers (ViTs), inspired by their success in natural language processing, have recently gained attention for heart sound classification (HSC). However, most of the existing studies on HSC rely on single-stream architectures, overlooking the advantages of multi-resolution features. While multi-stream architectures employing early [...] Read more.

Vision Transformers (ViTs), inspired by their success in natural language processing, have recently gained attention for heart sound classification (HSC). However, most of the existing studies on HSC rely on single-stream architectures, overlooking the advantages of multi-resolution features. While multi-stream architectures employing early or late fusion strategies have been proposed, they often fall short of effectively capturing cross-modal feature interactions. Additionally, conventional fusion methods, such as concatenation, averaging, or max pooling, frequently result in information loss. To address these limitations, this paper presents a novel attention fusion-based two-stream Vision Transformer (AFTViT) architecture for HSC that leverages two-dimensional mel-cepstral domain features. The proposed method employs a ViT-based encoder to capture long-range dependencies and diverse contextual information at multiple scales. A novel attention block is then used to integrate cross-context features at the feature level, enhancing the overall feature representation. Experiments conducted on the PhysioNet2016 and PhysioNet2022 datasets demonstrate that the AFTViT outperforms state-of-the-art CNN-based methods in terms of accuracy. These results highlight the potential of the AFTViT framework for early diagnosis of cardiovascular diseases, offering a valuable tool for cardiologists and researchers in developing advanced HSC techniques. Full article

(This article belongs to the Section Biosignal Processing)

► Show Figures

Figure 1

29 pages, 1340 KB

Open AccessArticle

Cognitively Inspired Federated Learning Framework for Interpretable and Privacy-Secured EEG Biomarker Prediction of Depression Relapse

by Sana Yasin, Umar Draz, Tariq Ali, Mohammad Hijji, Muhammad Ayaz, El-Hadi M. Aggoune and Isha Yasin

Bioengineering 2025, 12(10), 1032; https://doi.org/10.3390/bioengineering12101032 - 26 Sep 2025

Abstract

Depression relapse is a common issue during long-term care. We introduce a privacy-aware explainable personalized federated learning (PFL) framework that incorporates layer-wise relevance propagation and Shapley value analysis to provide patient-specific interpretable predictions from EEG. The study is conducted with the publicly available [...] Read more.

Depression relapse is a common issue during long-term care. We introduce a privacy-aware explainable personalized federated learning (PFL) framework that incorporates layer-wise relevance propagation and Shapley value analysis to provide patient-specific interpretable predictions from EEG. The study is conducted with the publicly available Healthy Brain Network (HBN) dataset, with analysis conducted for n = 100 subjects with resting-state 128-channel EEG with accompanying psychometric scores, and subject-wise 10-fold cross-validation is used to assess the performance of the model. Multi-channel EEG features and standardized symptom scales are jointly modeled to both increase the clinical context of the model and avoid leakage issues. This results in overall accuracy, precision, recall, and F1-score values of 92%, 91%, 93%, and 90.5%, respectively. The attribution maps from the model suggest region-anchored spectral patterns that are associated with relapse risk, providing clinical interpretability, and the federated setup of the model allows for a privacy-aware training setup that is more easily adaptable to multi-site deployment. Together, these results suggest a scalable and clinically feasible approach to trustworthy relapse monitoring with earlier intervention. Full article

(This article belongs to the Special Issue Machine Learning and Deep Learning Applications in Healthcare, 2nd Edition)

11 pages, 606 KB

Open AccessArticle

Performance of ChatGPT-4 as an Auxiliary Tool: Evaluation of Accuracy and Repeatability on Orthodontic Radiology Questions

by Mercedes Morales Morillo, Nerea Iturralde Fernández, Luis Daniel Pellicer Castillo, Ana Suarez, Yolanda Freire and Victor Diaz-Flores García

Bioengineering 2025, 12(10), 1031; https://doi.org/10.3390/bioengineering12101031 - 26 Sep 2025

Abstract

Background: Large language models (LLMs) are increasingly considered in dentistry, yet their accuracy in orthodontic radiology remains uncertain. This study evaluated the performance of ChatGPT-4 on questions aligned with current radiology guidelines. Methods: Fifty short, guideline-anchored questions were authored; thirty were pre-selected a [...] Read more.

Background: Large language models (LLMs) are increasingly considered in dentistry, yet their accuracy in orthodontic radiology remains uncertain. This study evaluated the performance of ChatGPT-4 on questions aligned with current radiology guidelines. Methods: Fifty short, guideline-anchored questions were authored; thirty were pre-selected a priori for their diagnostic relevance. Using the ChatGPT-4 web interface in March 2025, we obtained 30 answers per item (900 in total) across two user accounts and three times of day, each in a new chat with a standardised prompt. Two blinded experts graded all responses on a 3-point scale (0 = incorrect, 1 = partially correct, 2 = correct); disagreements were adjudicated. The primary outcome was strict accuracy (proportion of answers graded 2). Secondary outcomes were partial-credit performance (mean 0–2 score) and inter-rater agreement using multiple coefficients. Results: Strict accuracy was 34.1% (95% CI 31.0–37.2), with wide item-level variability (0–100%). The mean partial-credit score was 1.09/2.00 (median 1.02; IQR 0.53–1.83). Inter-rater agreement was high (percent agreement: 0.938, with coefficients indicating substantial to almost-perfect reliability). Conclusions: In the conditions of this study, ChatGPT-4 demonstrated limited strict accuracy yet substantial reliability in expert grading when applied to orthodontic radiology questions. These findings underline its potential as a complementary educational and decision-support resource while also highlight its present limitations. Its role should remain supportive and informative, never replacing the critical appraisal and professional judgement of the clinician. Full article

(This article belongs to the Special Issue AI-Powered Diagnosis and Treatment Plans in Dentistry and Orofacial Fields)

27 pages, 3413 KB

Open AccessArticle

DermaMamba: A Dual-Branch Vision Mamba Architecture with Linear Complexity for Efficient Skin Lesion Classification

by Zhongyu Yao, Yuxuan Yan, Zhe Liu, Tianhang Chen, Ling Cho, Yat-Wah Leung, Tianchi Lu, Wenjin Niu, Zhenyu Qiu, Yuchen Wang, Xingcheng Zhu and Ka-Chun Wong

Bioengineering 2025, 12(10), 1030; https://doi.org/10.3390/bioengineering12101030 - 26 Sep 2025

Abstract

Accurate skin lesion classification is crucial for the early detection of malignant lesions, including melanoma, as well as improved patient outcomes. While convolutional neural networks (CNNs) excel at capturing local morphological features, they struggle with global context modeling essential for comprehensive lesion assessment. [...] Read more.

Accurate skin lesion classification is crucial for the early detection of malignant lesions, including melanoma, as well as improved patient outcomes. While convolutional neural networks (CNNs) excel at capturing local morphological features, they struggle with global context modeling essential for comprehensive lesion assessment. Vision transformers address this limitation but suffer from quadratic computational complexity O(n²), hindering deployment in resource-constrained clinical environments. We propose DermaMamba, a novel dual-branch fusion architecture that integrates CNN-based local feature extraction with Vision Mamba (VMamba) for efficient global context modeling with linear complexity O(n). Our approach introduces a state space fusion mechanism with adaptive weighting that dynamically balances local and global features based on lesion characteristics. We incorporate medical domain knowledge through multi-directional scanning strategies and ABCDE (Asymmetry, Border irregularity, Color variation, Diameter, Evolution) rule feature integration. Extensive experiments on the ISIC dataset show that DermaMamba achieves 92.1% accuracy, 91.7% precision, 91.3% recall, and 91.5% mac-F1 score, which outperforms the best baseline by 2.0% accuracy with 2.3× inference speedup and 40% memory reduction. The improvements are statistically significant based on a significance test (p < 0.001, Cohen’s d > 0.8), with greater than 79% confidence also preserved on challenging boundary cases. These results establish DermaMamba as an effective solution bridging diagnostic accuracy and computational efficiency for clinical deployment. Full article

(This article belongs to the Special Issue AI and Machine Learning in Medical Image Processing: Innovations and Applications)

► Show Figures

Figure 1

18 pages, 3979 KB

Open AccessArticle

Hemodynamic Alteration in Aortic Valve Stenosis: CFD Insights from Leaflet-Resolved Models

by Mashrur Muntasir Nuhash, Victor K. Lai and Ruihang Zhang

Bioengineering 2025, 12(10), 1029; https://doi.org/10.3390/bioengineering12101029 - 26 Sep 2025

Abstract

Aortic valve stenosis, is a prevalent cardiovascular disease, narrows the valve orifice and restricts blood flow, resulting in abnormal high velocities and shear stresses. The progression of these hemodynamic abnormalities and their link with stenosis severity remain incompletely understood, which are critical for [...] Read more.

Aortic valve stenosis, is a prevalent cardiovascular disease, narrows the valve orifice and restricts blood flow, resulting in abnormal high velocities and shear stresses. The progression of these hemodynamic abnormalities and their link with stenosis severity remain incompletely understood, which are critical for early detection and intervention. Computational Fluid Dynamics (CFD) was employed to characterize aortic hemodynamics across healthy, mild, moderate, and severe stenosis using a 3D steady-state model with idealized leaflet geometries. Key flow parameters, including velocity distribution, wall shear stress (WSS), pressure loss coefficient, and helicity, were evaluated. Results show a non-linear increase in velocity and WSS with stenosis severity, with peak jet velocities of 1.08, 1.82, 2.73, and 4.7 m/s and peak WSS of 11, 35, 80, and 122 Pa at the aortic arch, respectively. Severe stenosis produced a highly eccentric jet along the anterior of aortic arch, accompanied by a narrower jet, increased turbulence intensity and expanded recirculation zones. A significant increase in helicity and pressure loss coefficient was also observed for higher stenosis severities. These findings highlight the influence of valve leaflets on aortic flow dynamics, providing physiologically relevant insights into stenosis-induced mechanical stresses that may drive endothelial dysfunction and support earlier detection of disease progression. Full article

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

► Show Figures

Figure 1

40 pages, 19754 KB

Open AccessArticle

Trans-cVAE-GAN: Transformer-Based cVAE-GAN for High-Fidelity EEG Signal Generation

by Yiduo Yao, Xiao Wang, Xudong Hao, Hongyu Sun, Ruixin Dong and Yansheng Li

Bioengineering 2025, 12(10), 1028; https://doi.org/10.3390/bioengineering12101028 - 26 Sep 2025

Abstract

Electroencephalography signal generation remains a challenging task due to its non-stationarity, multi-scale oscillations, and strong spatiotemporal coupling. Conventional generative models, including VAEs and GAN variants such as DCGAN, WGAN, and WGAN-GP, often yield blurred waveforms, unstable spectral distributions, or lack semantic controllability, limiting [...] Read more.

Electroencephalography signal generation remains a challenging task due to its non-stationarity, multi-scale oscillations, and strong spatiotemporal coupling. Conventional generative models, including VAEs and GAN variants such as DCGAN, WGAN, and WGAN-GP, often yield blurred waveforms, unstable spectral distributions, or lack semantic controllability, limiting their effectiveness in emotion-related applications. To address these challenges, this research proposes a Transformer-based conditional variational autoencoder–generative adversarial network (Trans-cVAE-GAN) that combines Transformer-driven temporal modeling, label-conditioned latent inference, and adversarial learning. A multi-dimensional structural loss further constrains generation by preserving temporal correlation, frequency-domain consistency, and statistical distribution. Experiments on three SEED-family datasets—SEED, SEED-FRA, and SEED-GER—demonstrate high similarity to real EEG, with representative mean ± SD correlations of Pearson ≈ 0.84 ± 0.08/0.74 ± 0.12/0.84 ± 0.07 and Spearman ≈ 0.82 ± 0.07/0.72 ± 0.12/0.83 ± 0.08, together with low spectral divergence (KL ≈ 0.39 ± 0.15/0.41 ± 0.20/0.37 ± 0.18). Comparative analyses show consistent gains over classical GAN baselines, while ablations verify the indispensable roles of the Transformer encoder, label conditioning, and cVAE module. In downstream emotion recognition, augmentation with generated EEG raises accuracy from 86.9% to 91.8% on SEED (with analogous gains on SEED-FRA and SEED-GER), underscoring enhanced generalization and robustness. These results confirm that the proposed approach simultaneously ensures fidelity, stability, and controllability across cohorts, offering a scalable solution for affective computing and brain–computer interface applications. 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

10 pages, 525 KB

Open AccessArticle

Class III Malocclusion in Growing Patients: Facemask vs. Functional Appliance: Experimental Study

by Lucia Giannini, Guido Galbiati, Cinzia Maspero, Gianna Dipalma and Roberto Biagi

Bioengineering 2025, 12(10), 1027; https://doi.org/10.3390/bioengineering12101027 - 26 Sep 2025

Abstract

Objective: We compared the skeletal effects of postero-anterior facemask (PAF) and functional appliance (FA) therapy in growing patients with Class III malocclusion. Materials and Methods: A total of 85 patients (mean age 9 ± 0.2 years) were treated with either a PAF (n [...] Read more.

Objective: We compared the skeletal effects of postero-anterior facemask (PAF) and functional appliance (FA) therapy in growing patients with Class III malocclusion. Materials and Methods: A total of 85 patients (mean age 9 ± 0.2 years) were treated with either a PAF (n = 50) or a FA (n = 35). Pre- and post-treatment cephalometric records were analyzed to assess sagittal (SNA, SNB, ANB, Wits appraisal) and vertical changes. Treatment outcomes were compared using Student’s t test for paired samples. Results: PAF therapy produced significantly greater improvements in the ANB angle (mean increase 4.1° vs. 1.7°) and Wits appraisal (2.4 mm vs. 0.9 mm) compared to FAs. Vertical control was superior in the PAF group, which showed a reduction in lower facial height, whereas FA patients exhibited a slight increase. Conclusions: PAF therapy was more effective than FAs in improving both sagittal and vertical skeletal relationships in growing Class III patients. Functional appliances provided only modest skeletal effects, mainly influencing mandibular position. Early intervention with a PAF should be considered the treatment of choice when maxillary protraction and vertical control are required. Full article

(This article belongs to the Special Issue New Sight for the Treatment of Dental Diseases: Updates and Direction)

► Show Figures

Figure 1

6 pages, 173 KB

Open AccessEditorial

AI Advancements in Healthcare: Medical Imaging and Sensing Technologies

by Mohammed A. Al-masni and Kanghyun Ryu

Bioengineering 2025, 12(10), 1026; https://doi.org/10.3390/bioengineering12101026 - 26 Sep 2025

Abstract

Artificial intelligence (AI), broadly defined as algorithms capable of self-learning patterns from large-scale data, has emerged as one of the most transformative technologies in modern healthcare [...] Full article

(This article belongs to the Special Issue AI Advancements in Healthcare: Medical Imaging and Sensing Technologies)

14 pages, 813 KB

Open AccessArticle

Can Artificial Intelligence Improve the Appropriate Use and Decrease the Misuse of REBOA?

by Mary Bokenkamp, Yu Ma, Ander Dorken-Gallastegi, Jefferson A. Proaño-Zamudio, Anthony Gebran, George C. Velmahos, Dimitris Bertsimas and Haytham M. A. Kaafarani

Bioengineering 2025, 12(10), 1025; https://doi.org/10.3390/bioengineering12101025 - 25 Sep 2025

Abstract

Background: The use of resuscitative endovascular balloon occlusion of the aorta (REBOA) for control of noncompressible torso hemorrhage remains controversial. We aimed to utilize a novel and transparent/interpretable artificial intelligence (AI) method called Optimal Policy Trees (OPTs) to improve the appropriate use and [...] Read more.

Background: The use of resuscitative endovascular balloon occlusion of the aorta (REBOA) for control of noncompressible torso hemorrhage remains controversial. We aimed to utilize a novel and transparent/interpretable artificial intelligence (AI) method called Optimal Policy Trees (OPTs) to improve the appropriate use and decrease the misuse of REBOA in hemodynamically unstable blunt trauma patients. Methods: We trained and then validated OPTs that “prescribe” REBOA in a 50:50 split on all hemorrhagic shock blunt trauma patients in the 2010–2019 ACS-TQIP database based on rates of survival. Hemorrhagic shock was defined as a systolic blood pressure ≤90 on arrival or a transfusion requirement of ≥4 units of blood in the first 4 h of presentation. The expected 24 h mortality rate following OPT prescription was compared to the observed 24 h mortality rate in patients who were or were not treated with REBOA. Results: Out of 4.5 million patients, 100,615 were included, and 803 underwent REBOA. REBOA patients had a higher rate of pelvic fracture, femur fracture, hemothorax, pneumothorax, and thoracic aorta injury (p < 0.001). The 24 h mortality rate for the REBOA vs. non-REBOA group was 47% vs. 21%, respectively (p < 0.001). OPTs resulted in an 18% reduction in 24 h mortality for REBOA and a 0.8% reduction in non-REBOA patients. We specifically divert the misuse of REBOA by recommending against REBOA in cases where it leads to worse outcomes. Conclusions: This proof-of-concept study shows that interpretable AI models can improve mortality in unstable blunt trauma patients by optimizing the use and decreasing the misuse of REBOA. To date, these models have been used to predict outcomes, but their groundbreaking use will be in prescribing interventions and changing outcomes. Full article

(This article belongs to the Section Biosignal Processing)

► Show Figures

Figure 1

42 pages, 14694 KB

Open AccessReview

Exploration of Glitazone/Thiazolidinedione Derivatives: Molecular Design and Therapeutic Potential

by Salahuddin, Avijit Mazumder, Mohamed Jawed Ahsan, Rajnish Kumar, Zabih Ullah, Mohammad Shahar Yar and Km Shabana

Bioengineering 2025, 12(10), 1024; https://doi.org/10.3390/bioengineering12101024 - 25 Sep 2025

Abstract

This review of thiazolidinedione or glitazone, which have a five-membered heterocyclic ring C₃NS, shows their versatile properties in terms of pharmacological actions such as antimicrobial, antifungal, insecticidal, pesticidal, antidiabetic, anti-inflammatory, anti-proliferative, anti-neurotoxicity, anticonvulsant, anti-thyroidal, and anti-tubercular uses. While having a wide [...] Read more.

This review of thiazolidinedione or glitazone, which have a five-membered heterocyclic ring C₃NS, shows their versatile properties in terms of pharmacological actions such as antimicrobial, antifungal, insecticidal, pesticidal, antidiabetic, anti-inflammatory, anti-proliferative, anti-neurotoxicity, anticonvulsant, anti-thyroidal, and anti-tubercular uses. While having a wide range of biological activities, the TZDs mainly act via binding to the peroxisome proliferator-activated receptor (PPAR) members. PPAR-γ are ligand-activated transcription factors, which are members of the nuclear hormone receptors group. Activations of PPAR-γ regulate cell proliferation and differentiation, glucose homeostasis, apoptosis, lipid metabolism, and inflammatory responses. This review explores the synthesis of a thiazolidinedione and its derivatives, focusing on their pharmacological profiles and antidiabetic activity. It highlights the benefits of synthesis, reaction profiles, and catalyst recovery, which may encourage further investigation into these scaffolds by researchers. Based on synthesized derivatives, some glimpses of the structure–activity relationships of some compounds have been compiled. All the synthesized derivatives have been reviewed concerning their standard drugs already available and concluded with the highly or moderately active synthesized derivatives of thiazolidinedione. The data for this review was collected by an extensive review of current scientific literature, including on the synthesis, biological evaluation, SAR, and patents (2015–25). Full article

(This article belongs to the Section Biochemical Engineering)

► Show Figures

Figure 1

14 pages, 2926 KB

Open AccessArticle

A Dual-Thread Lag–Locking Screw Enhances Single Lateral Plate Fixation in Bicondylar Tibial Plateau Fractures: A Biomechanical Study

by Ya-Han Chan, Hsuan-Wen Wang, Wei-Che Tsai and Chun-Li Lin

Bioengineering 2025, 12(10), 1023; https://doi.org/10.3390/bioengineering12101023 - 25 Sep 2025

Abstract

Schatzker type V bicondylar tibial plateau fractures present a major challenge due to the difficulty of achieving stable fixation with minimally invasive strategies. This study introduces a dual-thread lag and locking plate (DLLP) design that integrates lag screw compression with unilateral locking plate [...] Read more.

Schatzker type V bicondylar tibial plateau fractures present a major challenge due to the difficulty of achieving stable fixation with minimally invasive strategies. This study introduces a dual-thread lag and locking plate (DLLP) design that integrates lag screw compression with unilateral locking plate fixation. A custom-built compression evaluation platform and standardized 3D-printed fracture models were employed to assess biomechanical performance. DLLP produced measurable interfragmentary compression during screw insertion, with a mean displacement of 1.22 ± 0.11 mm compared with 0.02 ± 0.04 mm for conventional single lateral locking plates (SLLPs) (p < 0.05). In static testing, DLLP demonstrated a significantly greater maximum failure force (7801.51 ± 358.95 N) than SLLP (6224.84 ± 411.20 N, p < 0.05) and improved resistance to lateral displacement at 2 mm (3394.85 ± 392.81 N vs. 2766.36 ± 64.51 N, p = 0.03). Under dynamic fatigue loading simulating one year of functional use, all DLLP constructs survived 1 million cycles with <2 mm displacement, while all SLLP constructs failed prematurely (mean fatigue life: 408,679 ± 128,286 cycles). These findings highlight the critical role of lag screw compression in maintaining fracture stability and demonstrate that DLLP provides superior biomechanical performance compared with SLLP, supporting its potential as a less invasive alternative to dual plating in the treatment of complex tibial plateau fractures. Full article

(This article belongs to the Special Issue Orthopedic and Trauma Biomechanics)

► Show Figures

Figure 1

12 pages, 1025 KB

Open AccessArticle

Detecting Event-Related Spectral Perturbations in Right-Handed Sensorimotor Cortical Responses Using OPM-MEG

by Hao Lu, Yong Li, Min Xiang, Yuyu Ma, Yang Gao and Xiaolin Ning

Bioengineering 2025, 12(10), 1022; https://doi.org/10.3390/bioengineering12101022 - 25 Sep 2025

Abstract

The optically pumped magnetometer, OPM-MEG, has the potential to replace the traditional low-temperature superconducting quantum interference device, SQUID-MEG. Event-related spectral perturbations (ERSPs) can be used to examine the temporal- and frequency-domain characteristics of a signal. In this paper, a finger-tapping movement paradigm based [...] Read more.

The optically pumped magnetometer, OPM-MEG, has the potential to replace the traditional low-temperature superconducting quantum interference device, SQUID-MEG. Event-related spectral perturbations (ERSPs) can be used to examine the temporal- and frequency-domain characteristics of a signal. In this paper, a finger-tapping movement paradigm based on auditory cues is adopted, and OPM-MEG is used to measure the functional signals of the brain. The event-related spectral perturbation values of the right and left hands of right-handed people were calculated and compared. The results showed that there was a significant difference in the event-related spectral perturbations between the right and left hands of right-handed people. In summary, OPM-MEG has the ability to measure the event-related spectral perturbations of the brain during finger movements and verify the asymmetry of motor skills. Full article

(This article belongs to the Special Issue Electric, Magnetic, and Electromagnetic Fields in Biology and Medicine: From Mechanisms to Biomedical Applications: 3rd Edition)

► Show Figures

Figure 1

51 pages, 8270 KB

Open AccessReview

Advances of Functional Two-Dimensional Nanomaterials in the Treatment of Oral Diseases

by Ziyi Xu, Rong Meng, Yue Wang, Yuxuan Sun, Jiao Qiao, Yang Yao and Qiang Peng

Bioengineering 2025, 12(10), 1021; https://doi.org/10.3390/bioengineering12101021 - 25 Sep 2025

Abstract

Two-dimensional (2D) nanomaterials have attracted growing attention in the field of oral medicine due to their unique physicochemical properties, including high surface area, adjustable surface chemistry, and exceptional biocompatibility. In recent years, a variety of 2D materials, including graphene-based nanomaterials, black phosphorus nanosheets, [...] Read more.

Two-dimensional (2D) nanomaterials have attracted growing attention in the field of oral medicine due to their unique physicochemical properties, including high surface area, adjustable surface chemistry, and exceptional biocompatibility. In recent years, a variety of 2D materials, including graphene-based nanomaterials, black phosphorus nanosheets, MXenes, layered double hydroxides (LDHs), transition metal dichalcogenides (TMDs), 2D metal–organic frameworks (MOFs), and polymer-based nanosheets, have been extensively explored for the treatment of oral diseases. These functional materials demonstrate multiple therapeutic capabilities, such as antibacterial activity, reactive oxygen species (ROS) scavenging, anti-inflammatory modulation, and promotion of tissue regeneration. In this review, we systematically summarize the recent advances of 2D nanomaterials in the treatment of common oral diseases such as dental caries, periodontitis, oral cancer and peri-implantitis. The underlying therapeutic mechanisms are also summarized. Challenges for clinical translation of these nanomaterials and the possible solutions are discussed as well. Full article

(This article belongs to the Special Issue Nano–Bio Interface—Second Edition)

► Show Figures

Figure 1

18 pages, 725 KB

Open AccessArticle

Breast Cancer Prediction Using Rotation Forest Algorithm Along with Finding the Influential Causes

by Prosenjit Das, Proshenjit Sarker, Jun-Jiat Tiang and Abdullah-Al Nahid

Bioengineering 2025, 12(10), 1020; https://doi.org/10.3390/bioengineering12101020 - 25 Sep 2025

Abstract

Breast cancer is a widespread disease involving abnormal (uncontrolled) growth of breast tissue cells along with the formation of a tumor and metastasis. Breast cancer cases occur mostly among women. Early detection and regular screening have significantly improved survival rates. This research classifies [...] Read more.

Breast cancer is a widespread disease involving abnormal (uncontrolled) growth of breast tissue cells along with the formation of a tumor and metastasis. Breast cancer cases occur mostly among women. Early detection and regular screening have significantly improved survival rates. This research classifies breast cancer and non-breast cancer cases using machine learning algorithms based on the Breast Cancer Coimbra dataset by optimizing the classifier performance and feature selection methodology. In addition, this research identifies the influential features responsible for BC classification by using diverse counterfactual explanations. The Rotation Forest classifier algorithm is used to classify breast cancer and non-breast cancer cases. The hyperparameters of this algorithm are optimized using the Optuna optimizer. Three wrapper-based feature selection techniques (Sequential Forward Selection, Sequential Backward Selection, and Exhaustive Feature Selection) are used to select the most relevant features. An ensemble environment is also created using the best feature subsets of these methods, incorporating both soft and hard voting strategies. Experimental results show that the hard voting strategy achieves an accuracy of 85.71%, F1-score of 83.87%, precision of 92.85%, and recall of 76.47%. In contrast, the soft voting strategy obtains an accuracy of 80.00%, F1-score of 77.42%, precision of 85.71%, and recall of 70.59%. These findings demonstrate that hard voting achieves noticeably better performance. The misclassification outcomes of both strategies are explored using Diverse Counterfactual Explanations, revealing that BMI and Glucose values are most influential in predicting correct classes, whereas the HOMA, Adiponectin, and Resistin values have little influence. Full article

(This article belongs to the Section Biosignal Processing)

► Show Figures

Figure 1

24 pages, 2075 KB

Open AccessReview

Fibroblast Growth Factor-Derived Peptides: Sources, Functions, and Applications

by Cheng-Kun Cao, Zhong-Yuan Shi, Chuan-Bang Chen, Xiao-Kun Li and Zhi-Jian Su

Bioengineering 2025, 12(10), 1019; https://doi.org/10.3390/bioengineering12101019 - 25 Sep 2025

Abstract

Fibroblast growth factors (FGFs) play a crucial role in various biological processes, including tissue development, metabolic regulation, and injury repair. Previous studies have shown that certain peptides can exhibit similar biological functions to FGFs, whether they are fragments extracted from natural FGF molecules [...] Read more.

Fibroblast growth factors (FGFs) play a crucial role in various biological processes, including tissue development, metabolic regulation, and injury repair. Previous studies have shown that certain peptides can exhibit similar biological functions to FGFs, whether they are fragments extracted from natural FGF molecules or derived peptides designed based on the structural characteristics of FGFs and their receptor molecules. These FGF-derived peptides have shown significant application potential in fields including tissue repair and regeneration, cancer therapy, metabolic regulation, neural recovery, and biological delivery. This article reviews the sources, bioactive functions, molecular mechanisms, and application prospects of FGF-derived peptides, aiming to provide new research ideas for subsequent structural optimization, drug delivery system development, and clinical translation of these peptides. Full article

(This article belongs to the Special Issue Biologically Active Constituents and Their Applications for Skin Regeneration and Repair)

► Show Figures

Figure 1

15 pages, 678 KB

Open AccessArticle

Comparative Analysis of Knee Biomechanics in Total Knee Arthroplasty Patients Across Daily Activities

by Fangjian Chen, Hannah Seymour and Naiquan (Nigel) Zheng

Bioengineering 2025, 12(10), 1018; https://doi.org/10.3390/bioengineering12101018 - 25 Sep 2025

Abstract

Total knee arthroplasty (TKA) is a commonly conducted surgery to relieve pain and enhance mobility in patients with end-stage knee osteoarthritis. Patient-reported outcome measures are often used whereas biomechanical variables are too complicated for clinicians and patients to assess functional improvement. There is [...] Read more.

Total knee arthroplasty (TKA) is a commonly conducted surgery to relieve pain and enhance mobility in patients with end-stage knee osteoarthritis. Patient-reported outcome measures are often used whereas biomechanical variables are too complicated for clinicians and patients to assess functional improvement. There is a need for a simplified integrated knee biomechanics index (KBI) to compare improvements in TKA patients across various daily activities and examine the relationships between clinical functional tests and daily activities. Age-, gender-, and BMI-matched three groups (20 each in posterior stabilized TKA, bi-cruciate stabilized TKA, and healthy controls) were recruited and tested pre-op and 6-month post-op to perform walking on level, slope, and stairs, and two clinical tests (timed-up-go, 10-time sit-to-stand). Knee joint kinematics and kinetics variables were calculated from motion data and ground reactions captured at 120 Hz and 1200 Hz, respectively. KBI was developed based on these variables relative to healthy controls. The longitude comparison of KBI and the differences of KBI across various daily activities were identified using repeated-measure ANOVA. Pearson correlation analysis was used to compare clinical tests and KBI of daily activities. KBIs of five daily activities were significantly increased following TKA follow-up. KBI improvement during level walking was significantly higher than those during stair ascending and descending. Significant correlations were found between timed-up-go test time and KBIs for stair ascending and descending. Full article

(This article belongs to the Section Biomechanics and Sports Medicine)

► Show Figures

Figure 1

13 pages, 2523 KB

Open AccessArticle

Body Size Modulates the Impact of the Dispersive Patch Position During Radiofrequency Cardiac Ablation

by Ramiro M. Irastorza and Enrique Berjano

Bioengineering 2025, 12(10), 1017; https://doi.org/10.3390/bioengineering12101017 - 24 Sep 2025

Viewed by 9

Abstract

(1) Background: In the context of cardiac radiofrequency (RF) ablation, it has been proposed that positioning the dispersive patch (DP) concordantly with the orientation of the ablation electrode may enhance lesion size. The objective of this study is to investigate how individual body [...] Read more.

(1) Background: In the context of cardiac radiofrequency (RF) ablation, it has been proposed that positioning the dispersive patch (DP) concordantly with the orientation of the ablation electrode may enhance lesion size. The objective of this study is to investigate how individual body size may modulate the extent of this effect. (2) Methods: Three computational models representing different body sizes were developed. An irrigated catheter ablation was simulated by delivering a 30 W pulse for 30 s to the endocardial surface of the anterior wall. Lesion sizes were then compared between two configurations of the dispersive patch (DP): an anterior (concordant) position and a posterior (discordant) position. (3) Results: Lesion size was consistently and significantly greater with concordant DP positioning compared to discordant positioning. Moreover, the magnitude of this difference decreased significantly with increasing body size, ranging from 0.65 ± 0.08 mm in the 35 kg swine model to 0.51 ± 0.06 mm in the human model. (4) Conclusions: Body size has a modest influence on the effect of dispersive patch positioning on RF lesion size. The potential advantage of a concordant DP configuration may be more significant in individuals with smaller body volume. Full article

(This article belongs to the Special Issue Electromagnetic Fields in Bioengineering: Advancing the OneHealth Paradigm)

► Show Figures

Figure 1

28 pages, 4876 KB

Open AccessArticle

Evaluating the Molecular Basis of Nanocalcium-Induced Health Regulation in Zebra Fish (Danio rerio)

by Madhubala Kumari, Aastha Tiwary, Rishav Sheel, Arnab Roy Chowdhury, Biplab Sarkar, Koel Mukherjee and Dipak Maity

Bioengineering 2025, 12(10), 1016; https://doi.org/10.3390/bioengineering12101016 - 24 Sep 2025

Viewed by 12

Abstract

The present study aimed to evaluate the impact of varying dietary concentrations of calcium oxide nanoparticles (CaO-NPs) on important health regulators in Zebra fish (Danio rerio) using integrative physiological, histopathological, and computational approaches. The co-precipitation method was used to synthesize NPs and [...] Read more.

The present study aimed to evaluate the impact of varying dietary concentrations of calcium oxide nanoparticles (CaO-NPs) on important health regulators in Zebra fish (Danio rerio) using integrative physiological, histopathological, and computational approaches. The co-precipitation method was used to synthesize NPs and characterization was performed through DLS, XRD, FESEM, EDX, and FTIR depicting spherical-shaped CaO-NPs with a hydrodynamic diameter of 91.2 nm. Adult Danio rerio were administered with three different feed regimes enriched with 2.4 (T1), 1.6 (T2), and 0.8 (T3) mg CaO-NPs/kg for 30 days. Growth, survival, NP accumulation, and histological assessments, and bioinformatic studies, were performed to understand interactions of NPs with fish metabolic proteins. The T3 group demonstrated the highest survival (75%) and weight gain (+39.31%), and exhibited the lowest accumulation of CaO-NPs in the brain (0.133 mg/L), liver (0.642 mg/L), and intestine (0.773 mg/L) with no evident histological alterations, whereas T1 group exhibited major liver and intestinal damage. Molecular docking targeting the NRF-2 oxidative stress pathway revealed strong binding affinities of NPs with catalase (−3.7), keap1a (−3.5), keap1b (−3.3), and mafk (−2.4), highlighting potential modulation of redox homeostasis. Hence, a 0.8mg CaO-NPs/kg feed dose is recommended to promote potential health benefits in Danio rerio, which can be further applicable to commercial aquaculture for enhanced fish health while minimizing toxicity. Full article

(This article belongs to the Special Issue Nano–Bio Interface—Second Edition)

► Show Figures

Figure 1

16 pages, 258 KB

Open AccessReview

Focal Adhesion of Collagen-Based Bone Grafting Materials Enhances Bone Regeneration

by Mao-Suan Huang, Tzu-Sen Yang, Chia-Jung Wang, John F. Bowley and Wen-Fu T. Lai

Bioengineering 2025, 12(10), 1015; https://doi.org/10.3390/bioengineering12101015 - 24 Sep 2025

Viewed by 145

Abstract

Collagen, which has osteoconductive potential, has been widely used as a scaffold material for bone repair and regeneration for more than the last three decades. Recently, collagen has been combined with other materials to produce collagen-based bone grafting materials with enhanced bone repair [...] Read more.

Collagen, which has osteoconductive potential, has been widely used as a scaffold material for bone repair and regeneration for more than the last three decades. Recently, collagen has been combined with other materials to produce collagen-based bone grafting materials with enhanced bone repair and regeneration capacities. However, varied results have been obtained with collagen-based grafting materials. Methods: To elucidate the mechanisms underlying the enhanced bone engineering capacity of these materials, we critically reviewed the current literature on the complex hierarchical structure and properties of native collagen molecules. Results: This review highlights the scientific challenge of manufacturing collagen-based materials with suitable properties and shapes for specific biomedical applications, particularly those related to bone repair and regeneration. Conclusions: This article sheds light on the interactions between collagen and cell receptor molecules to mediate biological pathways. In addition, this article clarifies the mechanisms of cell adhesion-mediated bone regeneration. The findings may guide future research on collagen-based biomaterials. Full article

(This article belongs to the Special Issue Advances in Bone Regeneration with Multipotential Stromal Cells and Bioactive Materials)

15 pages, 1685 KB

Open AccessArticle

Ultra-High Resolution 9.4T Brain MRI Segmentation via a Newly Engineered Multi-Scale Residual Nested U-Net with Gated Attention

by Aryan Kalluvila, Jay B. Patel and Jason M. Johnson

Bioengineering 2025, 12(10), 1014; https://doi.org/10.3390/bioengineering12101014 - 24 Sep 2025

Viewed by 154

Abstract

A 9.4T brain MRI is the highest resolution MRI scanner in the public market. It offers submillimeter brain imaging with exceptional anatomical detail, making it one of the most powerful tools for detecting subtle structural changes associated with neurological conditions. Current segmentation models [...] Read more.

A 9.4T brain MRI is the highest resolution MRI scanner in the public market. It offers submillimeter brain imaging with exceptional anatomical detail, making it one of the most powerful tools for detecting subtle structural changes associated with neurological conditions. Current segmentation models are optimized for lower-field MRI (1.5T–3T), and they struggle to perform well on 9.4T data. In this study, we present the GA-MS-UNet++, the world’s first deep learning-based model specifically designed for 9.4T brain MRI segmentation. Our model integrates multi-scale residual blocks, gated skip connections, and spatial channel attention mechanisms to improve both local and global feature extraction. The model was trained and evaluated on 12 patients in the UltraCortex 9.4T dataset and benchmarked against four leading segmentation models (Attention U-Net, Nested U-Net, VDSR, and R2UNet). The GA-MS-UNet++ achieved a state-of-the-art performance across both evaluation sets. When tested against manual, radiologist-reviewed ground truth masks, the model achieved a Dice score of 0.93. On a separate test set using SynthSeg-generated masks as the ground truth, the Dice score was 0.89. Across both evaluations, the model achieved an overall accuracy of 97.29%, precision of 90.02%, and recall of 94.00%. Statistical validation using the Wilcoxon signed-rank test (p < 1 × 10⁻⁵) and Kruskal–Wallis test (H = 26,281.98, p < 1 × 10⁻⁵) confirmed the significance of these results. Qualitative comparisons also showed a near-exact alignment with ground truth masks, particularly in areas such as the ventricles and gray–white matter interfaces. Volumetric validation further demonstrated a high correlation (R² = 0.90) between the predicted and ground truth brain volumes. Despite the limited annotated data, the GA-MS-UNet++ maintained a strong performance and has the potential for clinical use. This algorithm represents the first publicly available segmentation model for 9.4T imaging, providing a powerful tool for high-resolution brain segmentation and driving progress in automated neuroimaging analysis. Full article

(This article belongs to the Special Issue New Sights of Machine Learning and Digital Models in Biomedicine)

► Show Figures

Figure 1

18 pages, 1015 KB

Open AccessArticle

Edge-Driven Disability Detection and Outcome Measurement in IoMT Healthcare for Assistive Technology

by Malak Alamri, Khalid Haseeb, Mamoona Humayun, Menwa Alshammeri, Ghadah Naif Alwakid and Naeem Ramzan

Bioengineering 2025, 12(10), 1013; https://doi.org/10.3390/bioengineering12101013 - 23 Sep 2025

Viewed by 127

Abstract

The integration of edge computing (EC) and Internet of Medical Things (IoMT) technologies facilitates the development of adaptive healthcare systems that significantly improve the accessibility and monitoring of individuals with disabilities. By enabling real-time disease identification and reducing response times, this architecture supports [...] Read more.

The integration of edge computing (EC) and Internet of Medical Things (IoMT) technologies facilitates the development of adaptive healthcare systems that significantly improve the accessibility and monitoring of individuals with disabilities. By enabling real-time disease identification and reducing response times, this architecture supports personalized healthcare solutions for those with chronic conditions or mobility impairments. The inclusion of untrusted devices leads to communication delays and enhances the security risks for medical applications. Therefore, this research presents a Trust-Driven Disability-Detection Model Using Secured Random Forest Classification (TTDD-SRF) to address the issues while monitoring real-time health records. It also increases the detection of abnormal movement patterns to highlight the indication of disability using edge-driven communication. The TTDD-SRF model improves the classification accuracy of abnormal motion detection while ensuring data reliability through trust scores computed at the edge level. Such a paradigm decreases the ratio of false positives and enhances decision-making accuracy in coping with health-related applications, mainly the detection of patients’ disabilities. The experimental analysis of the proposed TTDD-SRF model indicates improved performance in terms of network throughput by 48%, system resilience by 42%, device integrity by 49%, and energy consumption by 45% while highlighting the potential of medical systems using edge technologies, advancing assistive technology for healthcare accessibility. Full article

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

► Show Figures

Figure 1

20 pages, 2924 KB

Open AccessArticle

Feature Selection and Prediction of Pediatric Tuina in Attention Deficit/Hyperactivity Disorder Management: A Machine Learning Approach Based on Parent-Reported Children’s Constitution

by Shu-Cheng Chen, Guo-Tao Wu, Han Li, Xuan Zhang, Zi-Han Li, Pong-Ming Wong, Le-Fei Han, Jing Qin, Kwai-Ching Lo, Wing-Fai Yeung and Ge Ren

Bioengineering 2025, 12(10), 1012; https://doi.org/10.3390/bioengineering12101012 - 23 Sep 2025

Viewed by 131

Abstract

Background: Attention Deficit/Hyperactivity Disorder (ADHD) is a common neurodevelopmental disorder in children. Pediatric tuina, a traditional Chinese medicine (TCM) intervention, has shown potential in managing ADHD symptoms. Integrating machine learning (ML) into pediatric tuina could refine treatment personalization, allowing for a [...] Read more.

Background: Attention Deficit/Hyperactivity Disorder (ADHD) is a common neurodevelopmental disorder in children. Pediatric tuina, a traditional Chinese medicine (TCM) intervention, has shown potential in managing ADHD symptoms. Integrating machine learning (ML) into pediatric tuina could refine treatment personalization, allowing for a more feasible and better parent-administered use. Methods: We employed an ML-based model to analyze parent-reported constitutional features from 1005 children diagnosed with ADHD to predict individualized pediatric tuina treatments. This study focused on feature selection and the application of several ML models, including Support Vector Machines (SVM), Logistic Regression (LR), Multilayer Perceptron (MLP), and Random Forest (RF). The key task involved identifying the most relevant features for effective TCM pattern identification and diagnosis, which would guide personalized treatment strategies. Results: The ML models displayed strong predictive performance, with the MLP model achieving the highest Area Under the Curve (AUC) of 0.90 and an accuracy (ACC) of 0.74. Seven features were selected five times in cross-validation. This facilitated a more targeted and effective pediatric tuina application tailored to individual constitution. Conclusion: This study developed an ML-based approach to enhance ADHD management in children using pediatric tuina, informed by a parent-reported questionnaire. It identified seven key features for TCM pattern identification and personalized treatment strategies. MLP achieved the highest AUC and ACC. Full article

(This article belongs to the Special Issue Deep Learning in Medical Applications: Challenges and Opportunities)

► Show Figures

Figure 1

29 pages, 8257 KB

Open AccessArticle

Characterization of a Novel POx-Based Adhesive Powder for Obliterating Dead Spaces After Surgery

by Steven E. M. Poos, Roger M. L. M. Lomme, Edwin A. Roozen, Johan C. M. E. Bender, Harry van Goor and Richard P. G. Ten Broek

Bioengineering 2025, 12(10), 1011; https://doi.org/10.3390/bioengineering12101011 - 23 Sep 2025

Viewed by 180

Abstract

Surgical dead spaces are challenging to handle with current preventive methods. Tissue adhesives show promise in obliterating ‘dead spaces’, but the drawbacks of currently available adhesives prevent them from being used for dead space elimination. An adhesive powder based on N-Hydroxysuccinimide-poly(2-oxazoline), NHS-POx, combines [...] Read more.

Surgical dead spaces are challenging to handle with current preventive methods. Tissue adhesives show promise in obliterating ‘dead spaces’, but the drawbacks of currently available adhesives prevent them from being used for dead space elimination. An adhesive powder based on N-Hydroxysuccinimide-poly(2-oxazoline), NHS-POx, combines robust adhesive strength in moist environments with favorable biocompatibility and biodegradability, which makes this an interesting candidate for eliminating spaces that remain between tissues after surgery. The current study evaluates the swelling, crosslinking speed, and degradation properties of this novel tissue adhesive. These results were then used to design multiple adhesive variants differing in pH, surfactant addition, and particle size, which were subsequently examined based on their wetting rates, adhesive strength, and durability. The powder displayed minimal swelling and rapid crosslinking properties, by which the latter could be increased by a basic buffer or surfactant addition and reduced by increasing particle size. The wetting rate of the powder increased when a surfactant (Pluronic F68) was added to the mix. The adhesive strength, as measured by tensile and shear strength measurements of different prototypes of the adhesive powder, was significantly better than that of a commercially available fibrin glue. The addition of both buffer and Pluronic F68 led to a breakdown of adhesive force after 14 days of incubation, while the prototype containing neither buffer nor Pluronic F68 still had measurable adhesive force after 14 days of incubation. The current study results display several characteristics of the NHS-POx-based tissue adhesive that are favorable for tissue approximation, preventing the occurrence of dead spaces. The most effective and usable adhesive prototype will be identified in further ex vivo and in vivo animal model studies. Full article

(This article belongs to the Special Issue Advanced Biomaterials, Coatings, and Techniques: Applications in Medicine)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Bioengineering, Volume 12, Issue 10 (October 2025) – 33 articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI