Bioengineering

22 pages, 1271 KiB

Open AccessArticle

OKEN: A Supervised Evolutionary Optimizable Dimensionality Reduction Framework for Whole Slide Image Classification

by Soroush Oskouei, André Pedersen, Marit Valla, Vibeke Grotnes Dale, Sissel Gyrid Freim Wahl, Mats Dehli Haugum, Thomas Langø, Maria Paula Ramnefjell, Lars Andreas Akslen, Gabriel Kiss and Hanne Sorger

Bioengineering 2025, 12(7), 733; https://doi.org/10.3390/bioengineering12070733 (registering DOI) - 4 Jul 2025

Abstract

Classification of lung cancer subtypes is a critical clinical step; however, relying solely on H&E-stained histopathology images can pose challenges, and additional immunohistochemical analysis is sometimes required for definitive subtyping. Digital pathology facilitates the use of artificial intelligence for automatic classification of digital [...] Read more.

Classification of lung cancer subtypes is a critical clinical step; however, relying solely on H&E-stained histopathology images can pose challenges, and additional immunohistochemical analysis is sometimes required for definitive subtyping. Digital pathology facilitates the use of artificial intelligence for automatic classification of digital tissue slides. Automatic classification of Whole Slide Images (WSIs) typically involves extracting features from patches obtained from them. The aim of this study was to develop a WSI classification framework utilizing an optimizable kernel to encode features from each patch from a WSI into a desirable and adjustable latent space using an evolutionary algorithm. The encoded data can then be used for classification and segmentation while being computationally more efficient. Our proposed framework is compared with a state-of-the-art model, Vim4Path, on an internal and external dataset of lung cancer WSIs. The proposed model outperforms Vim-S16 in accuracy and F₁ score at both

\times 2.5

and

\times 10

magnification levels on the internal test set, with the highest accuracy (0.833) and F₁ score (0.721) at

\times 2.5

. On the external test set, Vim-S16 at

\times 10

achieves the highest accuracy (0.732), whereas OKEN-DenseNet121 at

\times 2.5

has the best F₁ score (0.772). In future work, finding a dynamic way to tune the output dimensions of the evolutionary algorithm would be of value. Full article

(This article belongs to the Special Issue Application of Deep Learning in Medical Diagnosis)

► Show Figures

Figure 1

16 pages, 7005 KiB

Open AccessArticle

Comparison of Tissue Repair with Different Types of Microdissection Tips: A Randomized Histomorphometric Evaluation in Rats

by Ana Luiza Vila Verde Brunelli, Luíz Henrique Soares Torres, Arthur Henrique Alécio Viotto, Izabela Fornazari Delamura, Ana Paula Farnezi Bassi, Marisa Aparecida Cabrini Gabrielli and Valfrido Antonio Pereira-Filho

Bioengineering 2025, 12(7), 732; https://doi.org/10.3390/bioengineering12070732 (registering DOI) - 4 Jul 2025

Abstract

The aim of the study was to compare tissue repair of incisions made using different microdissection electrocautery tips in an in vivo animal model. Skin incisions were made, including the subcutaneous tissue, in 30 adult Wistar rats using four types of instruments: a [...] Read more.

The aim of the study was to compare tissue repair of incisions made using different microdissection electrocautery tips in an in vivo animal model. Skin incisions were made, including the subcutaneous tissue, in 30 adult Wistar rats using four types of instruments: a scalpel blade number 15, knife-type electrocautery, microdissection needle, and thin-cut electrode. The animals were divided into five groups based on the euthanasia time—24 h, 48 h, 72 h, 7 days, and 14 days. Each animal received four incisions, one with each type of instrument. Histological and histomorphometric analyses were performed using hematoxylin and eosin (HE) and Picrosirius red stains. Analysis of variance (ANOVA) showed that the type of dissector had no significant effect on type I collagen levels (p = 0.615), whereas the euthanasia time had a significant influence (p < 0.001). Estimated marginal means for type I collagen showed minimal variation among groups, ranging from 35.4% to 36.5%, suggesting limited clinical differences between instruments. These results indicate that while the choice of dissector has a limited impact on type I collagen deposition, time is a determining factor in the wound healing process. The thin-cut electrode enables incisions with tissue repair comparable to that of a number 15 scalpel, as it performs cutting, coagulation, and blending functions at lower temperatures. Full article

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

► Show Figures

Figure 1

29 pages, 2934 KiB

Open AccessSystematic Review

Cardiac Simulator Technologies and Design for Medical Education and Auscultation Training: A Systematic Review

by Christian Romero-Martínez, Luis Adrián Zúñiga-Avilés, Giorgio M. Cruz-Martínez, José Javier Reyes-Lagos, Joel Zagoya-López and Ángel Eduardo Bárcenas-García

Bioengineering 2025, 12(7), 731; https://doi.org/10.3390/bioengineering12070731 - 3 Jul 2025

Abstract

Medical simulators have revolutionized clinical training, particularly in teaching skills such as cardiac auscultation. This review synthesizes recent advances in the technological design and implementation of cardiac simulators for medical education, alongside scientometric and patentometric analyses. The focus is on innovations enhancing efficacy, [...] Read more.

Medical simulators have revolutionized clinical training, particularly in teaching skills such as cardiac auscultation. This review synthesizes recent advances in the technological design and implementation of cardiac simulators for medical education, alongside scientometric and patentometric analyses. The focus is on innovations enhancing efficacy, safety, and accessibility. Analyses included 69 patents published over the past five years, sourced from Google Patents, Patentscope, Espacenet, and The Lens. A bibliometric analysis was performed using 52 scientific reports from PubMed, ScienceDirect, and The Lens within the same timeframe. Key findings indicate an 8% increase in AI-integrated cardiac auscultation devices compared to conventional equipment. Furthermore, 85% of the studies reported compliance with applicable regulations of at least 90%, reflecting improved regulatory alignment. This analysis provides a foundation for future research and the development of more accurate and accessible educational tools for cardiac auscultation training. Full article

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

► Show Figures

Figure 1

17 pages, 2255 KiB

Open AccessArticle

Predicting Fetal Growth with Curve Fitting and Machine Learning

by Huan Zhang, Chuan-Sheng Hung, Chun-Hung Richard Lin, Hong-Ren Yu, You-Cheng Zheng, Cheng-Han Yu, Chih-Min Tsai and Ting-Hsin Huang

Bioengineering 2025, 12(7), 730; https://doi.org/10.3390/bioengineering12070730 - 3 Jul 2025

Abstract

Monitoring fetal growth throughout pregnancy is essential for early detection of developmental abnormalities. This study developed a Taiwan-specific fetal growth reference using a web-based data collection platform and polynomial regression modeling. We analyzed ultrasound data from 980 pregnant women, encompassing 8350 prenatal scans, [...] Read more.

Monitoring fetal growth throughout pregnancy is essential for early detection of developmental abnormalities. This study developed a Taiwan-specific fetal growth reference using a web-based data collection platform and polynomial regression modeling. We analyzed ultrasound data from 980 pregnant women, encompassing 8350 prenatal scans, to model six key fetal biometric parameters: abdominal circumference, crown–rump length, estimated fetal weight, head circumference, biparietal diameter, and femur length. Quadratic regression was selected based on a balance of performance and simplicity, with R² values exceeding 0.95 for most parameters. Confidence intervals and real-time anomaly detection were implemented through the platform. The results demonstrate the potential for efficient, population-specific fetal growth monitoring in clinical settings. Full article

(This article belongs to the Special Issue Computer Vision and Machine Learning in Medical Applications, 2nd Edition)

► Show Figures

Figure 1

13 pages, 1243 KiB

Open AccessArticle

Three-Dimensional Assessment of the Biological Periacetabular Defect Reconstruction in an Ovine Animal Model: A µ-CT Analysis

by Frank Sebastian Fröschen, Thomas Martin Randau, El-Mustapha Haddouti, Jacques Dominik Müller-Broich, Frank Alexander Schildberg, Werner Götz, Dominik John, Susanne Reimann, Dieter Christian Wirtz and Sascha Gravius

Bioengineering 2025, 12(7), 729; https://doi.org/10.3390/bioengineering12070729 - 3 Jul 2025

Abstract

The increasing number of acetabular revision total hip arthroplasties requires the evaluation of alternative materials in addition to established standards using a defined animal experimental defect that replicates the human acetabular revision situation as closely as possible. Defined bone defects in the load-bearing [...] Read more.

The increasing number of acetabular revision total hip arthroplasties requires the evaluation of alternative materials in addition to established standards using a defined animal experimental defect that replicates the human acetabular revision situation as closely as possible. Defined bone defects in the load-bearing area of the acetabulum were augmented with various materials in an ovine periacetabular defect model (Group 1: NanoBone^® (artificial hydroxyapatite-silicate composite; Artoss GmbH, Germany); Group 2: autologous sheep cancellous bone; Group 3: Tutoplast^® (processed allogeneic sheep cancellous bone; Tutogen Medical GmbH, Germany)) and bridged with an acetabular reinforcement ring of the Ganz type. Eight months after implantation, a μ-CT examination (n = 8 animals per group) was performed. A μ-CT analysis of the contralateral acetabula (n = 8, randomly selected from all three groups) served as the control group. In a defined volume of interest (VOI), bone volume (BV), mineral volume (MV), and bone substitute volume (BSV), as well as the bone surface (BS) relative to the total volume (TV) and the surface-to-volume ratio (BS/BV), were determined. To assess the bony microarchitecture, trabecular thickness (Tb.Th), trabecular separation (Tb.Sp), and trabecular number (Tb.N), as well as connectivity density (Conn.D), the degree of anisotropy (DA), and the structure model index (SMI), were evaluated. The highest BV was observed for NanoBone^® (Group 1), which also showed the highest proportion of residual bone substitute material in the defect. This resulted in a significant increase in BV/TV with a significant decrease in BS/BV. The assessment of the microstructure for Groups 2 and 3 compared to Group 1 showed a clear approximation of Tb.Th, Tb.Sp, Tb.N, and Conn.D to the microstructure of the control group. The SMI showed a significant decrease in Group 1. All materials demonstrated their suitability by supporting biological defect reconstruction. NanoBone^® showed the highest rate of new bone formation; however, the microarchitecture indicated more advanced bone remodeling and an approximate restoration of the trabecular structure for both autologous and allogeneic Tutoplast^® cancellous bone when using the impaction bone grafting technique. Full article

(This article belongs to the Special Issue Innovative Biomaterials and Advanced Technologies in Bone Tissue Regeneration)

► Show Figures

Figure 1

22 pages, 4293 KiB

Open AccessArticle

Speech-Based Parkinson’s Detection Using Pre-Trained Self-Supervised Automatic Speech Recognition (ASR) Models and Supervised Contrastive Learning

by Hadi Sedigh Malekroodi, Nuwan Madusanka, Byeong-il Lee and Myunggi Yi

Bioengineering 2025, 12(7), 728; https://doi.org/10.3390/bioengineering12070728 - 1 Jul 2025

Abstract

Diagnosing Parkinson’s disease (PD) through speech analysis is a promising area of research, as speech impairments are often one of the early signs of the disease. This study investigates the efficacy of fine-tuning pre-trained Automatic Speech Recognition (ASR) models, specifically Wav2Vec 2.0 and [...] Read more.

Diagnosing Parkinson’s disease (PD) through speech analysis is a promising area of research, as speech impairments are often one of the early signs of the disease. This study investigates the efficacy of fine-tuning pre-trained Automatic Speech Recognition (ASR) models, specifically Wav2Vec 2.0 and HuBERT, for PD detection using transfer learning. These models, pre-trained on large unlabeled datasets, can be capable of learning rich speech representations that capture acoustic markers of PD. The study also proposes the integration of a supervised contrastive (SupCon) learning approach to enhance the models’ ability to distinguish PD-specific features. Additionally, the proposed ASR-based features were compared against two common acoustic feature sets: mel-frequency cepstral coefficients (MFCCs) and the extended Geneva minimalistic acoustic parameter set (eGeMAPS) as a baseline. We also employed a gradient-based method, Grad-CAM, to visualize important speech regions contributing to the models’ predictions. The experiments, conducted using the NeuroVoz dataset, demonstrated that features extracted from the pre-trained ASR models exhibited superior performance compared to the baseline features. The results also reveal that the method integrating SupCon consistently outperforms traditional cross-entropy (CE)-based models. Wav2Vec 2.0 and HuBERT with SupCon achieved the highest F1 scores of 90.0% and 88.99%, respectively. Additionally, their AUC scores in the ROC analysis surpassed those of the CE models, which had comparatively lower AUCs, ranging from 0.84 to 0.89. These results highlight the potential of ASR-based models as scalable, non-invasive tools for diagnosing and monitoring PD, offering a promising avenue for the early detection and management of this debilitating condition. Full article

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

► Show Figures

Figure 1

12 pages, 1162 KiB

Open AccessArticle

Coronal Alignment Does Not Adequately Predict Femoral Rotation Axes in Total Knee Arthroplasty: Application of a 3D Image-Based Robotic-Assisted Arthroplasty Platform

by Utkarsh Anil, Catherine Di Gangi, Lachlan Anderson, Charles C. Lin, Matthew Hepinstall, Morteza Meftah and Armin Arshi

Bioengineering 2025, 12(7), 727; https://doi.org/10.3390/bioengineering12070727 - 1 Jul 2025

Abstract

(1) Introduction: Precise femoral component rotation is critical for achieving symmetric flexion-gap balance and physiologic patellofemoral tracking in mechanically aligned total knee arthroplasty (TKA). Surgeons often infer an appropriate rotational target from the patient’s coronal limb alignment, yet the strength of this relationship [...] Read more.

(1) Introduction: Precise femoral component rotation is critical for achieving symmetric flexion-gap balance and physiologic patellofemoral tracking in mechanically aligned total knee arthroplasty (TKA). Surgeons often infer an appropriate rotational target from the patient’s coronal limb alignment, yet the strength of this relationship remains uncertain. (2) Methods: We identified 695 consecutive patients undergoing primary TKA with a preoperative planning CT scan. The surgical transepicondylar axis (sTEA) and posterior condylar axis (PCAxis) were identified and the angle between them was measured. The angle between the mechanical axis of the femur and tibia was used to measure the coronal alignment of the limb. (3) Results: The mean sTEA was 3.0° externally rotated to the PCAxis (range 3.1° internal to 9.2° external). The mean coronal alignment was 4.3° varus (range −12.5° valgus to 24.5° varus). There were 465 patients with >2° varus and 101 patients with >2° valgus. The mean sTEA was 2.9 ± 1.9° externally rotated relative to the PCAxis in the valgus group and 2.8 ± 2.0° in the varus group, with no statistically significant difference (p = 0.7). (4) Conclusions: There is significant variation in the femoral rotation axes between patients, but no significant relationship between overall limb coronal alignment and the magnitude of femoral rotation axes variation. This reinforces the need for independent assessment of rotational landmarks when performing mechanically aligned TKA. Full article

(This article belongs to the Special Issue Total Joint Arthroplasty: Technical Developments and Applications, 2nd Edition)

► Show Figures

Figure 1

17 pages, 3833 KiB

Open AccessArticle

Exercises Based on a Non-Immersive Virtual Reality Application for Upper Limb Rehabilitation

by Cosmin-Ilie Cotia and Silviu Dan Mandru

Bioengineering 2025, 12(7), 726; https://doi.org/10.3390/bioengineering12070726 - 1 Jul 2025

Abstract

Virtual reality (VR) technologies have gained increasing attention in the field of physical rehabilitation due to their potential to enhance patient engagement and provide adaptive, feedback-rich environments. In this study, we report on the development and preliminary evaluation of a VR-based rehabilitation application [...] Read more.

Virtual reality (VR) technologies have gained increasing attention in the field of physical rehabilitation due to their potential to enhance patient engagement and provide adaptive, feedback-rich environments. In this study, we report on the development and preliminary evaluation of a VR-based rehabilitation application aimed at improving upper extremity function, including muscle strength, endurance, and joint mobility. The application delivers a structured set of interactive exercises designed to support recovery through engaging, gamified tasks with real-time performance feedback and scalable difficulty levels. A pilot usability study was conducted with a cohort of target users to assess the system’s practicality, therapeutic relevance, and user satisfaction. Qualitative data were collected to evaluate usability, effectiveness, and areas for further improvement. Preliminary results suggest that the VR application is usable, accessible, and well-received by users, with high levels of engagement reported throughout the intervention. Participants also provided constructive feedback, emphasizing the potential benefits of incorporating enhanced sensory feedback mechanisms to improve immersion and therapeutic impact. These initial findings support the viability of VR-based rehabilitation tools and provide a foundation for future clinical studies aimed at validating their efficacy in diverse patient populations. Full article

(This article belongs to the Special Issue Augmented Reality, Virtual Reality and Extended Reality Solutions for Biomedical Applications)

► Show Figures

Figure 1

3 pages, 604 KiB

Open AccessEditorial

Arboviruses and COVID-19: Global Health Challenges and Human Enhancement Technologies

by Rengarajan Murugesan, Srinivasan Prabhu, John T. D. Caleb, Yuvaraj Maria Francis and Indra Neel Pulidindi

Bioengineering 2025, 12(7), 725; https://doi.org/10.3390/bioengineering12070725 - 1 Jul 2025

Abstract

Currently, there is increasing concern about severe illnesses, particularly those caused by viruses such as SARS-CoV-2 and arboviruses [...] Full article

► Show Figures

Figure 1

11 pages, 3678 KiB

Open AccessArticle

Plug-and-Play Self-Supervised Denoising for Pulmonary Perfusion MRI

by Changyu Sun, Yu Wang, Cody Thornburgh, Ai-Ling Lin, Kun Qing, John P. Mugler III and Talissa A. Altes

Bioengineering 2025, 12(7), 724; https://doi.org/10.3390/bioengineering12070724 - 1 Jul 2025

Abstract

Pulmonary dynamic contrast-enhanced (DCE) MRI is clinically useful for assessing pulmonary perfusion, but its signal-to-noise ratio (SNR) is limited. A self-supervised learning network-based plug-and-play (PnP) denoising model was developed to improve the image quality of pulmonary perfusion MRI. A dataset of patients with [...] Read more.

Pulmonary dynamic contrast-enhanced (DCE) MRI is clinically useful for assessing pulmonary perfusion, but its signal-to-noise ratio (SNR) is limited. A self-supervised learning network-based plug-and-play (PnP) denoising model was developed to improve the image quality of pulmonary perfusion MRI. A dataset of patients with suspected pulmonary diseases was used. Asymmetric pixel-shuffle downsampling blind-spot network (AP-BSN) training inputs were two-dimensional background-subtracted perfusion images without clean ground truth. The AP-BSN is incorporated into a PnP model (PnP-BSN) for balancing noise control and image fidelity. Model performance was evaluated by SNR, sharpness, and overall image quality from two radiologists. The fractal dimension and k-means segmentation of the pulmonary perfusion images were calculated for comparing denoising performance. The model was trained on 29 patients and tested on 8 patients. The performance of PnP-BSN was compared to denoising convolutional neural network (DnCNN) and a Gaussian filter. PnP-BSN showed the highest reader scores in terms of SNR, sharpness, and overall image quality as scored by two radiologists. The expert scoring results for DnCNN, Gaussian, and PnP-BSN were 2.25 ± 0.65, 2.44 ± 0.73, and 3.56 ± 0.73 for SNR; 2.62 ± 0.52, 2.62 ± 0.52, and 3.38 ± 0.64 for sharpness; and 2.16 ± 0.33, 2.34 ± 0.42, and 3.53 ± 0.51 for overall image quality (p < 0.05 for all). PnP-BSN outperformed DnCNN and a Gaussian filter for denoising pulmonary perfusion MRI, which led to improved quantitative fractal analysis. Full article

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

► Show Figures

Figure 1

13 pages, 1733 KiB

Open AccessArticle

Towards an Automated Computational Workflow to Assess Primary Stability in Total Hip Arthroplasty

by Massimiliano Mercuri, Enrico Toccaceli, Xiaoshu Sun, Giuseppe Marongiu, Marco Viceconti, Antonino Amedeo La Mattina and Cristina Curreli

Bioengineering 2025, 12(7), 723; https://doi.org/10.3390/bioengineering12070723 - 30 Jun 2025

Abstract

Total hip arthroplasty is one of the most common and rapidly growing surgical procedures, with over one million cases performed annually in the United States. Despite high success rates, revision surgeries remain a significant concern due to complications such as aseptic loosening, often [...] Read more.

Total hip arthroplasty is one of the most common and rapidly growing surgical procedures, with over one million cases performed annually in the United States. Despite high success rates, revision surgeries remain a significant concern due to complications such as aseptic loosening, often resulting from inadequate primary implant stability. This study presents an automated computational framework that integrates three-dimensional preoperative planning and finite element modeling to predict the primary stability of hip implants. Data obtained from the virtual surgery phase are used to generate subject-specific finite element models, which are executed on high-performance computing systems. The simulation evaluates implant stability by analyzing the contact interaction between the bone and the implant. The pipeline is demonstrated using data from the open-source HFValid collection and a commercial implant. Automation substantially reduced the time required to set up simulations, improving the efficiency on high-performance infrastructure. This integrated computational approach bridges the gap between biomechanical modeling and clinical decision-making and can serve as a preclinical tool for identifying personalized implant strategies and for conducting large-scale virtual cohort studies. Full article

(This article belongs to the Special Issue Diagnostic Tools and Therapeutic Strategies for Hip Diseases)

► Show Figures

Figure 1

32 pages, 9748 KiB

Open AccessArticle

Construction of a Structurally Unbiased Brain Template with High Image Quality from MRI Scans of Saudi Adult Females

by Noura Althobaiti, Kawthar Moria, Lamiaa Elrefaei, Jamaan Alghamdi and Haythum Tayeb

Bioengineering 2025, 12(7), 722; https://doi.org/10.3390/bioengineering12070722 - 30 Jun 2025

Abstract

In brain mapping, structural templates derived from population-specific MRI scans are essential for normalizing individual brains into a common space. This normalization facilitates accurate group comparisons and statistical analyses. Although templates have been developed for various populations, none currently exist for the Saudi [...] Read more.

In brain mapping, structural templates derived from population-specific MRI scans are essential for normalizing individual brains into a common space. This normalization facilitates accurate group comparisons and statistical analyses. Although templates have been developed for various populations, none currently exist for the Saudi population. To our knowledge, this work introduces the first structural brain template constructed and evaluated from a homogeneous subset of T1-weighted MRI scans of 11 healthy Saudi female subjects aged 25 to 30. Our approach combines the symmetric model construction (SMC) method with a covariance-based weighting scheme to mitigate bias caused by over-represented anatomical features. To enhance the quality of the template, we employ a patch-based mean-shift intensity estimation method that improves image sharpness, contrast, and robustness to outliers. Additionally, we implement computational optimizations, including parallelization and vectorized operations, to increase processing efficiency. The resulting template exhibits high image quality, characterized by enhanced sharpness, improved tissue contrast, reduced sensitivity to outliers, and minimized anatomical bias. This Saudi-specific brain template addresses a critical gap in neuroimaging resources and lays a reliable foundation for future studies on brain structure and function in this population. Full article

(This article belongs to the Section Biosignal Processing)

► Show Figures

Figure 1

13 pages, 1152 KiB

Open AccessArticle

Machine Learning Models Derived from [¹⁸F]FDG PET/CT for the Prediction of Recurrence in Patients with Thymomas

by Angelo Castello, Luigi Manco, Margherita Cattaneo, Riccardo Orlandi, Lorenzo Rosso, Giorgio Alberto Croci, Luigia Florimonte, Giovanni Scribano, Alessandro Turra, Stefano Ferrero, Mario Nosotti, Gianpaolo Carrafiello, Massimo Castellani and Paolo Mendogni

Bioengineering 2025, 12(7), 721; https://doi.org/10.3390/bioengineering12070721 - 30 Jun 2025

Abstract

Background/Objectives: This study aimed to develop machine learning (ML) models to predict recurrence in thymoma patients using conventional and radiomic signatures extracted from preoperative [¹⁸F]FDG PET/CT. Methods: A total of 50 patients (25 males, 25 females; mean age 63.3 ± 14.2 [...] Read more.

Background/Objectives: This study aimed to develop machine learning (ML) models to predict recurrence in thymoma patients using conventional and radiomic signatures extracted from preoperative [¹⁸F]FDG PET/CT. Methods: A total of 50 patients (25 males, 25 females; mean age 63.3 ± 14.2 years) who underwent thymectomy and preoperative [¹⁸F]FDG PET/CT between 2012 and 2022 were retrospectively analyzed. Radiomic analysis was performed using free-from-recurrence (FFR) status as a reference. Clinico-metabolic PET parameters were collected, and thymoma lesions were manually segmented on [¹⁸F]FDG PET/CT. A total of 856 radiomic features (RFts) were extracted from PET and CT datasets following IBSI guidelines, and robust RFts were selected. The dataset was split into training (70%) and validation (30%) sets. Two ML models (PET- and CT-based, respectively), each with three classifiers—Random Forest (RF), Support-Vector-Machine, and Tree—were trained and internally validated using RFts and clinico-metabolic signatures. Results: A total of 50 ROIs were selected and segmented. FFR was observed in 84% of our cohort. Forty-three robust RFts were selected from the CT dataset and 16 from the PET dataset, predominantly wavelet-based RFts. Additionally, three metabolic PET parameters were selected and included in the PET Model. Both the CT and PET models successfully discriminated against FFR after surgery, with the CT Model slightly outperforming the PET Model across different classifiers. The performance metrics of the RF classifier for the CT and PET models were AUC = 0.970/0.949, CA = 0.880/0.840, Precision = 0.884/0.842, Recall = 0.880/0.846, Specificity = 0.887/0.839, Sensitivity = 0.920/0.844, TP = 81.8%/83.3%, and TN = 92.9%/84.6%, respectively. Conclusions: ML-models trained on PET/CT radiomic features show promising results for predicting recurrence in patients with thymomas, which could be potentially applied in clinical practice for a better personalized treatment strategy. Full article

(This article belongs to the Special Issue Application of Artificial Intelligence in Oncologic PET Imaging)

► Show Figures

Figure 1

11 pages, 8944 KiB

Open AccessArticle

High-Speed Full-Color Polarized Light Imaging of Collagen Using a Polarization Camera

by Bin Yang, Neil Nayyar and Billy Sanchez

Bioengineering 2025, 12(7), 720; https://doi.org/10.3390/bioengineering12070720 - 30 Jun 2025

Abstract

Polarized light imaging (PLI) has been effective in visualizing and quantifying collagen content. Collagen-specific data are often overlaid over the tissue image for visualization. However, such contextual tissue images are typically in grayscale and lack important color information, limiting the usefulness of PLI [...] Read more.

Polarized light imaging (PLI) has been effective in visualizing and quantifying collagen content. Collagen-specific data are often overlaid over the tissue image for visualization. However, such contextual tissue images are typically in grayscale and lack important color information, limiting the usefulness of PLI in imaging the stained histology slides and for surgical guidance. The objective of this study was to develop a robust and easy-to-implement PLI technique to capture both true color and birefringent collagen data, and we call it ColorPOL. ColorPOL uses only one polarization-sensitive camera to capture information at 75 frames per second. The true color images were synthesized from individual RGB images, and collagen-specific information (fiber orientation and retardance) was derived from the green channel image. We implemented ColorPOL in transmission mode on an upright microscope and in reflection mode for wide-field thick tissue imaging. The color images in both implementations provided valuable color tissue context that facilitated the identification and localization of collagen content. Additionally, we demonstrated that in reflection mode, the high imaging speed enabled us to record and visualize continuous deformations of the collagenous tissues (tendons, sciatic nerves, and blood vessels) overlaid on the processed collagen-specific information. Robust performance and flexible configuration will make ColorPOL a valuable tool in basic research and translational applications. Full article

(This article belongs to the Special Issue Recent Advances in Soft Tissue Characterization Using Optical Techniques)

► Show Figures

Figure 1

11 pages, 1033 KiB

Open AccessArticle

Correlation Between the Orientation of the Nasopalatine Canal and the Upper Central Incisors: A Comparative Radiological Study Using Cbct

by Alessia Lasaracina, Roberto Luongo, Sergio Aliota, Alessia Acquaviva, Calogero Bugea, Mariano Lauriola, Erda Qorri, Antonio Scarano and Sang-Choon Cho

Bioengineering 2025, 12(7), 719; https://doi.org/10.3390/bioengineering12070719 - 30 Jun 2025

Abstract

During implant placement in the upper anterior region, the nasopalatine canal (NPC) is a frequently encountered anatomical structure. It connects the nasal and oral cavities and contains critical blood vessels and nerves. Despite its clinical relevance, no study has yet assessed the orientation [...] Read more.

During implant placement in the upper anterior region, the nasopalatine canal (NPC) is a frequently encountered anatomical structure. It connects the nasal and oral cavities and contains critical blood vessels and nerves. Despite its clinical relevance, no study has yet assessed the orientation of the NPC in relation to the upper central incisors to aid in optimal implant positioning. This study investigated the parallelism between the NPC and the upper central incisors (1.1 and 2.1) in both the mesiodistal and buccopalatal directions. Data were collected from 226 subjects, and statistical analyses included Pearson’s correlation test, a one-sample t-test, and scatter plot analysis. The mean mesiodistal inclinations of the NPC, 1.1, and 2.1 were 87.54° ± 3.20, 86.55° ± 3.97, and 86.50° ± 3.63, respectively, while their buccopalatal inclinations measured 67.92° ± 6.89, 67.02° ± 6.88, and 67.23° ± 7.76, respectively. These findings indicate a strong correlation between the buccopalatal and mesiodistal inclinations of the NPC and the upper central incisors, with no significant differences observed. These results align with the existing literature on the anatomical variability and clinical significance of the NPC. This correlation suggests that evaluating the spatial relationship between the NPC and adjacent teeth could enhance implant surgery planning, leading to improved clinical outcomes and minimizing complications such as hemorrhage or paresthesia. Full article

(This article belongs to the Special Issue Computed Tomography for Oral and Maxillofacial Applications)

► Show Figures

Figure 1

13 pages, 2397 KiB

Open AccessReview

Plastic Reconstruction of Upper Extremity Defects in Necrotizing Soft Tissue Infections

by Karren M. Takamura and Jason J. Yoo

Bioengineering 2025, 12(7), 718; https://doi.org/10.3390/bioengineering12070718 - 30 Jun 2025

Abstract

Soft tissue reconstruction in patients with upper extremity necrotizing soft tissue infections (NSTIs) can be challenging; these defects can be large with exposed critical structures. Following appropriate source control and debridement, soft tissue reconstruction is based on size, exposed structures, medical co-morbidities and [...] Read more.

Soft tissue reconstruction in patients with upper extremity necrotizing soft tissue infections (NSTIs) can be challenging; these defects can be large with exposed critical structures. Following appropriate source control and debridement, soft tissue reconstruction is based on size, exposed structures, medical co-morbidities and the physiologic status of the patient. There are multiple options for soft tissue coverage from local wound care to free tissue transfer. Dermal substitutes can help prepare a healthy wound bed that can later accept a skin graft. Local rotational flaps, distant pedicled flaps and free flaps are also options depending on the patient and the defect. Patients can have good functional outcomes after soft tissue reconstruction after upper extremity NSTI. Full article

(This article belongs to the Special Issue Surgical Wound Infections and Management)

► Show Figures

Figure 1

14 pages, 1457 KiB

Open AccessArticle

Validation of Automated Somatotype Estimation Proposal Using Full-Body 3D Scanning

by Bibiána Ondrejová, Lucia Bednarčíková, Norbert Ferenčík and Jozef Živčák

Bioengineering 2025, 12(7), 717; https://doi.org/10.3390/bioengineering12070717 - 30 Jun 2025

Abstract

Somatotyping is essential for assessing body composition in sports science, anthropology, and medicine. Traditional methods, such as the Heath–Carter approach, rely on manual measurements, which can be prone to errors and variability. This study evaluates the validity and reliability of 3D body scanning [...] Read more.

Somatotyping is essential for assessing body composition in sports science, anthropology, and medicine. Traditional methods, such as the Heath–Carter approach, rely on manual measurements, which can be prone to errors and variability. This study evaluates the validity and reliability of 3D body scanning as an alternative to manual somatotyping. A total of 117 participants (49 males, 68 females) aged 18 to 27 years were assessed using both traditional anthropometric methods and a full-body 3D scanning system (TC² NX-16). The three somatotype components (ectomorphy, mesomorphy, and endomorphy) were calculated using the Heath–Carter method. A custom-developed application processed the scanned data to compute somatotype values. The results were compared using statistical metrics, including intraclass correlation coefficients (ICCs) and Bland–Altman analysis. The 3D scanning method showed high agreement (87.18%) with manual measurements. Minor discrepancies were observed particularly in the endomorphic component, which was slightly overestimated by 3D scanning. Mesomorphic and ectomorphic components exhibited minimal differences. Statistical analyses confirmed strong reliability with ICC values exceeding 0.87. Conclusions: Full-body 3D scanning is a viable, non-invasive, and efficient alternative to traditional somatotyping methods. Despite minor differences in endomorphy estimation, the overall accuracy and reliability supports its use in sports science, health monitoring, and anthropometric research. Future studies should refine predictive models for endomorphy estimation and integrate AI-driven classification techniques to enhance precision. Full article

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

► Show Figures

Figure 1

18 pages, 2325 KiB

Open AccessArticle

Ultrasound Improves Gallbladder Contraction Function: A Non-Invasive Experimental Validation Using Small Animals

by Run Guo, Tian Chen, Fan Ding, Li-Ping Liu, Fang Chen, Gang Zhao and Bo Zhang

Bioengineering 2025, 12(7), 716; https://doi.org/10.3390/bioengineering12070716 - 30 Jun 2025

Abstract

Background: Gallbladder hypomotility is a key pathogenic factor in cholelithiasis. Non-invasive interventions to enhance gallbladder contractility remain limited. Ultrasound therapy has shown promise in various muscular disorders, but its effects on gallbladder function are unexplored. Methods: This study employed low-intensity pulsed ultrasound (LIPUS) [...] Read more.

Background: Gallbladder hypomotility is a key pathogenic factor in cholelithiasis. Non-invasive interventions to enhance gallbladder contractility remain limited. Ultrasound therapy has shown promise in various muscular disorders, but its effects on gallbladder function are unexplored. Methods: This study employed low-intensity pulsed ultrasound (LIPUS) at a 3 MHz frequency and 0.8 W/cm² intensity with a 20% duty cycle to irradiate the gallbladder region of fasting guinea pigs. Gallbladder contractile function was evaluated through multiple complementary approaches: in vivo assessment via two-dimensional/three-dimensional ultrasound imaging to monitor volumetric changes; quantitative functional evaluation using nuclear medicine scintigraphy (^99mTc-HIDA); and ex vivo experiments including isolated gallbladder muscle strip tension measurements, histopathological analysis, α-smooth muscle actin (α-SMA) immunohistochemistry, and intracellular calcium fluorescence imaging. Results: Ultrasound significantly enhanced gallbladder emptying, evidenced by the volume reduction and increased ejection fraction. Scintigraphy confirmed accelerated bile transport in treated animals. Ex vivo analyses demonstrated augmented contractile force, amplitude, and frequency in ultrasound-treated smooth muscle. Histological examination revealed smooth muscle hypertrophy, α-SMA upregulation, and elevated intracellular calcium levels. Extended ultrasound exposure produced sustained functional improvements without tissue damage. Conclusions: Ultrasound effectively enhances gallbladder contractile function through mechanisms involving smooth muscle structural modification and calcium signaling modulation. These findings establish the experimental foundation for ultrasound as a promising non-invasive therapeutic approach to improve gallbladder motility and potentially prevent gallstone formation. Full article

(This article belongs to the Special Issue Innovations in Hollow Viscera Imaging: Engineering Breakthroughs and Clinical Translation)

► Show Figures

Figure 1

21 pages, 1903 KiB

Open AccessArticle

Unlocking Superior MFH Performance Below Hergt’s Biological Safety Limit: SPION-Based Magnetic Nanoplatforms Deliver High Heating Efficiency at Low AMF

by Atul Sudame and Dipak Maity

Bioengineering 2025, 12(7), 715; https://doi.org/10.3390/bioengineering12070715 - 30 Jun 2025

Abstract

Superparamagnetic iron oxide nanoparticles (SPIONs) have gained significant attention for Magnetic Fluid Hyperthermia (MFH)-based cancer therapy. However, achieving high heating efficiency under a biologically safe Alternating Magnetic Field (AMF) remains a challenge. This study investigates the synthesis and optimization of SPIONs encapsulated in [...] Read more.

Superparamagnetic iron oxide nanoparticles (SPIONs) have gained significant attention for Magnetic Fluid Hyperthermia (MFH)-based cancer therapy. However, achieving high heating efficiency under a biologically safe Alternating Magnetic Field (AMF) remains a challenge. This study investigates the synthesis and optimization of SPIONs encapsulated in TPGS-stabilized PLGA nanoparticles (TPS-NPs) using a modified single emulsion solvent evaporation (M-SESE) method. The aim was to achieve efficient magnetic heating under biologically safe AMF conditions while maintaining biocompatibility and colloidal stability, making these magnetic nanoplatforms suitable for MFH-based cancer treatment. TPS-NPs were characterized using various techniques, including Dynamic Light Scattering (DLS), Atomic Force Microscopy (AFM), Transmission Electron Microscopy (TEM), and Superconducting Quantum Interference Device (SQUID) magnetometry, to evaluate their hydrodynamic size (Dh), zeta potential (ζ), encapsulation efficiency, and superparamagnetic properties. Calorimetric MFH studies demonstrated superior heating efficiency, with Specific Absorption Rate (SAR) and Intrinsic Loss Power (ILP) values optimized at an AMF of 4.1 GAm⁻¹s⁻¹, remaining within Hergt’s biological safety limit (~5 GAm⁻¹s⁻¹). These findings suggest that SPION-encapsulated TPS-NPs exhibit enhanced heat induction, making them promising candidates for MFH-based cancer therapy. The study highlights their potential as multifunctional nanoplatforms for magnetic hyperthermia therapy, paving the way for clinical translation in oncology for advanced cancer treatment. Full article

(This article belongs to the Special Issue Emerging Nanomaterials and Nanotechnologies for Biomedical and Related Applications)

► 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