Bioengineering

21 pages, 4785 KiB

Open AccessArticle

A Unified Multi-Task Learning Model with Joint Reverse Optimization for Simultaneous Skin Lesion Segmentation and Diagnosis

by Mohammed A. Al-masni, Abobakr Khalil Al-Shamiri, Dildar Hussain and Yeong Hyeon Gu

Bioengineering 2024, 11(11), 1173; https://doi.org/10.3390/bioengineering11111173 - 20 Nov 2024

Cited by 1 | Viewed by 1216

Abstract

Classifying and segmenting skin cancer represent pivotal objectives for automated diagnostic systems that utilize dermoscopy images. However, these tasks present significant challenges due to the diverse shape variations of skin lesions and the inherently fuzzy nature of dermoscopy images, including low contrast and [...] Read more.

Classifying and segmenting skin cancer represent pivotal objectives for automated diagnostic systems that utilize dermoscopy images. However, these tasks present significant challenges due to the diverse shape variations of skin lesions and the inherently fuzzy nature of dermoscopy images, including low contrast and the presence of artifacts. Given the robust correlation between the classification of skin lesions and their segmentation, we propose that employing a combined learning method holds the promise of considerably enhancing the performance of both tasks. In this paper, we present a unified multi-task learning strategy that concurrently classifies abnormalities of skin lesions and allows for the joint segmentation of lesion boundaries. This approach integrates an optimization technique known as joint reverse learning, which fosters mutual enhancement through extracting shared features and limiting task dominance across the two tasks. The effectiveness of the proposed method was assessed using two publicly available datasets, ISIC 2016 and PH², which included melanoma and benign skin cancers. In contrast to the single-task learning strategy, which solely focuses on either classification or segmentation, the experimental findings demonstrated that the proposed network improves the diagnostic capability of skin tumor screening and analysis. The proposed method achieves a significant segmentation performance on skin lesion boundaries, with Dice Similarity Coefficients (DSC) of 89.48% and 88.81% on the ISIC 2016 and PH² datasets, respectively. Additionally, our multi-task learning approach enhances classification, increasing the F1 score from 78.26% (baseline ResNet50) to 82.07% on ISIC 2016 and from 82.38% to 85.50% on PH². This work showcases its potential applicability across varied clinical scenarios. Full article

(This article belongs to the Section Biosignal Processing)

► Show Figures

Graphical abstract

11 pages, 1225 KiB

Open AccessArticle

AI-Driven Prediction of Symptom Trajectories in Cancer Care: A Deep Learning Approach for Chemotherapy Management

by Joseph Finkelstein, Aref Smiley, Christina Echeverria and Kathi Mooney

Bioengineering 2024, 11(11), 1172; https://doi.org/10.3390/bioengineering11111172 - 20 Nov 2024

Cited by 2 | Viewed by 1401

Abstract

This study presents an advanced method for predicting symptom escalation in chemotherapy patients using Long Short-Term Memory (LSTM) networks and Convolutional Neural Networks (CNNs). The accurate prediction of symptom escalation is critical in cancer care to enable timely interventions and improve symptom management [...] Read more.

This study presents an advanced method for predicting symptom escalation in chemotherapy patients using Long Short-Term Memory (LSTM) networks and Convolutional Neural Networks (CNNs). The accurate prediction of symptom escalation is critical in cancer care to enable timely interventions and improve symptom management to enhance patients’ quality of life during treatment. The analytical dataset consists of daily self-reported symptom logs from chemotherapy patients, including a wide range of symptoms, such as nausea, fatigue, and pain. The original dataset was highly imbalanced, with approximately 84% of the data containing no symptom escalation. The data were resampled into varying interval lengths to address this imbalance and improve the model’s ability to detect symptom escalation (n = 3 to n = 7 days). This allowed the model to predict significant changes in symptom severity across these intervals. The results indicate that shorter intervals (n = 3 days) yielded the highest overall performance, with the CNN model achieving an accuracy of 81%, precision of 87%, recall of 80%, and an F1 score of 83%. This was an improvement over the LSTM model, which had an accuracy of 79%, precision of 85%, recall of 79%, and an F1 score of 82%. The model’s accuracy and recall declined as the interval length increased, though precision remained relatively stable. The findings demonstrate that both CNN’s temporospatial feature extraction and LSTM’s temporal modeling effectively capture escalation patterns in symptom progression. By integrating these predictive models into digital health systems, healthcare providers can offer more personalized and proactive care, enabling earlier interventions that may reduce symptom burden and improve treatment adherence. Ultimately, this approach has the potential to significantly enhance the overall quality of life for chemotherapy patients by providing real-time insights into symptom trajectories and guiding clinical decision making. Full article

► Show Figures

Figure 1

19 pages, 6477 KiB

Open AccessArticle

First- vs. Second-Generation Autologous Platelet Concentrates and Their Implications for Wound Healing: Differences in Proteome and Secretome

by Hanna L. Stiller, Natarajan Perumal, Caroline Manicam, Emily R. Trzeciak, Julia Todt, Kerstin Jurk, Andrea Tuettenberg, Sven Schumann, Eik Schiegnitz and Sebastian Blatt

Bioengineering 2024, 11(11), 1171; https://doi.org/10.3390/bioengineering11111171 - 20 Nov 2024

Cited by 1 | Viewed by 1182

Abstract

Differences in cell count and growth factor expression between first- and second-generation autologous platelet concentrates (APCs) have been well described. The debate over which formula best supports wound healing in various surgical procedures is still ongoing. This study aims to assess the whole [...] Read more.

Differences in cell count and growth factor expression between first- and second-generation autologous platelet concentrates (APCs) have been well described. The debate over which formula best supports wound healing in various surgical procedures is still ongoing. This study aims to assess the whole proteome assembly, cell content, immunological potential and pro-angiogenic potential of second-generation APC, Platelet-Rich Fibrin (PRF) vs. first-generation APC, Platelet-Rich Plasma (PRP). The global proteome of the APCs was analyzed using nano-liquid chromatography mass spectrometry. Blood cell concentrations were determined by an automated cell counter. The effect of APCs on macrophage polarization was analyzed by flow cytometry. A yolk sac membrane (YSM) assay was used to monitor the neo-vessel formation and capillary branching in vivo. Cell count analysis revealed a higher number/concentration of leukocytes in PRF vs. PRP. Incubation of macrophages with PRP or platelet-free plasma (PFP) did not induce a significant pro-inflammatory state but led to a shift to the M0/M2 phenotype as seen in wound healing for all tested formulas. Label-free proteomics analysis identified a total of 387 proteins from three biological replicates of the respective designated groups. PRF induced increased formation of neo-vessels and branching points in vivo in comparison to PRP and PFP (each p < 0.001), indicating the enhanced pro-angiogenic potential of PRF. Overall, PRF seems superior to PRP, an important representative of first-generation formulas. Inclusion of leucocytes in PRF compared to PRP suggested rather an anti-inflammatory effect on macrophages. These results are important to support the versatile clinical applications in regenerative medicine for second-generation autologous platelet concentrates to optimize wound healing. Full article

► Show Figures

Figure 1

17 pages, 4587 KiB

Open AccessArticle

Improving Brain Metabolite Detection with a Combined Low-Rank Approximation and Denoising Diffusion Probabilistic Model Approach

by Yeong-Jae Jeon, Kyung Min Nam, Shin-Eui Park and Hyeon-Man Baek

Bioengineering 2024, 11(11), 1170; https://doi.org/10.3390/bioengineering11111170 - 20 Nov 2024

Viewed by 982

Abstract

In vivo proton magnetic resonance spectroscopy (MRS) is a noninvasive technique for monitoring brain metabolites. However, it is challenged by a low signal-to-noise ratio (SNR), often necessitating extended scan times to compensate. One of the conventional techniques for noise reduction is signal averaging, [...] Read more.

In vivo proton magnetic resonance spectroscopy (MRS) is a noninvasive technique for monitoring brain metabolites. However, it is challenged by a low signal-to-noise ratio (SNR), often necessitating extended scan times to compensate. One of the conventional techniques for noise reduction is signal averaging, which is inherently time-consuming and can lead to participant discomfort, thus posing limitations in clinical settings. This study aimed to develop a hybrid denoising strategy that integrates low-rank approximation and denoising diffusion probabilistic model (DDPM) to enhance MRS data quality and shorten scan times. Using publicly available ¹H MRS datasets from 15 subjects, we applied the Casorati SVD and DDPM to obtain baseline and functional data during a pain stimulation task. This method significantly improved SNR, resulting in outcomes comparable to or better than averaging over 32 signals. It also provided the most consistent metabolite measurements and adequately tracked temporal changes in glutamate levels, correlating with pain intensity ratings after heating. These findings demonstrate that our approach enhances MRS data quality, offering a more efficient alternative to conventional methods and expanding the potential for the real-time monitoring of neurochemical changes. This contribution has the potential to advance MRS techniques by integrating advanced denoising methods to increase the acquisition speed and enhance the precision of brain metabolite analyses. Full article

(This article belongs to the Special Issue Monitoring and Analysis of Human Biosignals, 3rd Edition)

► Show Figures

Figure 1

16 pages, 11648 KiB

Open AccessArticle

Analysis of Umbilical Artery Hemodynamics in Development of Intrauterine Growth Restriction Using Computational Fluid Dynamics with Doppler Ultrasound

by Xue Song, Jingying Wang, Ke Sun and Chunhian Lee

Bioengineering 2024, 11(11), 1169; https://doi.org/10.3390/bioengineering11111169 - 20 Nov 2024

Viewed by 1159

Abstract

Intrauterine growth restriction (IUGR), the failure of the fetus to achieve his/her growth potential, is a common and complex problem in pregnancy. Clinically, IUGR is usually monitored using Doppler ultrasound of the umbilical artery (UA). The Doppler waveform is generally divided into three [...] Read more.

Intrauterine growth restriction (IUGR), the failure of the fetus to achieve his/her growth potential, is a common and complex problem in pregnancy. Clinically, IUGR is usually monitored using Doppler ultrasound of the umbilical artery (UA). The Doppler waveform is generally divided into three typical patterns in IUGR development, from normal blood flow (Normal), to the loss of end diastolic blood flow (LDBF), and even to the reversal of end diastolic blood flow (RDBF). Unfortunately, Doppler ultrasound hardly provides complete UA hemodynamics in detail, while the present in silico computational fluid dynamics (CFD) can provide this with the necessary ultrasound information. In this paper, CFD is employed to simulate the periodic UA blood flow for three typical states of IUGR, which shows comprehensive information on blood flow velocity, pressure, and wall shear stress (WSS). A new finding is the “hysteresis effect” between the UA blood flow velocity and pressure drop in which the former always changes after the latter by 0.1–0.2 times a cardiac cycle due to the unsteady flow. The degree of hysteresis is a promising indicator characterizing the evolution of IUGR. CFD successfully shows the hemodynamic details in different development situations of IUGR, and undoubtedly, its results would also help clinicians to further understand the relationship between the UA blood flow status and fetal growth restriction. Full article

(This article belongs to the Section Biosignal Processing)

► Show Figures

Figure 1

3 pages, 153 KiB

Open AccessEditor’s ChoiceEditorial

Computational Fluid Dynamics in Medicine and Biology

by Amirtahà Taebi

Bioengineering 2024, 11(11), 1168; https://doi.org/10.3390/bioengineering11111168 - 20 Nov 2024

Viewed by 1313

Abstract

This Special Issue of Bioengineering presents cutting-edge research on the applications of computational fluid dynamics (CFD) in medical and biological contexts [...] Full article

(This article belongs to the Special Issue Computational Fluid Dynamics in Medicine and Biology)

17 pages, 3033 KiB

Open AccessArticle

Proteoglycans Enhance the Therapeutic Effect of BMSC Transplantation on Osteoarthritis

by Chunxiao Ran, Tianhao Liu, Yongming Bao, Weidan Wang, Dongling Xue, Guangxiao Yin, Xiuzhi Zhang and Dewei Zhao

Bioengineering 2024, 11(11), 1167; https://doi.org/10.3390/bioengineering11111167 - 20 Nov 2024

Viewed by 1161

Abstract

Background: The injection of bone mesenchymal stem cells (BMSCs) for osteoarthritis (OA) treatment fails to address the disrupted extracellular microenvironment, limiting the differentiation and paracrine functions of BMSCs and resulting in suboptimal therapeutic outcomes. Proteoglycans (PGs) promote cell differentiation, tissue repair, and microenvironment [...] Read more.

Background: The injection of bone mesenchymal stem cells (BMSCs) for osteoarthritis (OA) treatment fails to address the disrupted extracellular microenvironment, limiting the differentiation and paracrine functions of BMSCs and resulting in suboptimal therapeutic outcomes. Proteoglycans (PGs) promote cell differentiation, tissue repair, and microenvironment remodeling. This study investigated the potential of combining PGs with BMSCs to increase the efficacy of OA treatment. Methods: We evaluated the effects of PG on BMSC and chondrocyte functions by adding various PG concentrations to the culture media. Additionally, a Transwell system was used to assess the impact of PG on the communication between BMSCs and chondrocytes. The results of the in vitro experiment were verified by tissue staining and immunohistochemistry following the treatment of OA model rats. Results: Our findings indicate that PG effectively induces Col II expression in BMSCs and enhances the paracrine secretion of TGF-β1, thereby activating the TGF-β signaling pathway in chondrocytes and increasing PRG4 gene expression. Compared with the other groups, the BMSC/PG treatment group presented a smoother articular surface and more robust extracellular matrix than the other groups in vivo, with significantly increased expression and distribution of Smad2/3 and PRG4. Conclusions: PG enhances BMSC differentiation into chondrocytes and stimulates paracrine TGF-β1 secretion. Proteoglycans not only promote chondrocyte differentiation and paracrine TGF-β1 signaling in BMSCs but also increase the sensitivity of chondrocytes to TGF-β1 secreted from BMSCs, leading to PRG4 expression through the TGFR/Smad2/3 pathway. Proteoglycans can enhance the therapeutic effect of BMSC treatment on OA and have the potential to delay the degeneration of OA cartilage. Full article

(This article belongs to the Special Issue Integrating Novel Therapeutics and Personalized Medicine into Osteoarthritis Treatment: Advancements in Cartilage Regeneration and Joint Preservation)

► Show Figures

Figure 1

1 pages, 170 KiB

Open AccessCorrection

Correction: Lana et al. Sacral Bioneuromodulation: The Role of Bone Marrow Aspirate in Spinal Cord Injuries. Bioengineering 2024, 11, 461

by José Fábio Lana, Annu Navani, Madhan Jeyaraman, Napoliane Santos, Luyddy Pires, Gabriel Silva Santos, Izair Jefthé Rodrigues, Douglas Santos, Tomas Mosaner, Gabriel Azzini, Lucas Furtado da Fonseca, Alex Pontes de Macedo, Stephany Cares Huber, Daniel de Moraes Ferreira Jorge and Joseph Purita

Bioengineering 2024, 11(11), 1166; https://doi.org/10.3390/bioengineering11111166 - 19 Nov 2024

Viewed by 632

Abstract

In the original publication [...] Full article

(This article belongs to the Special Issue Innovations in Nerve Regeneration)

13 pages, 3184 KiB

Open AccessReview

A Comprehensive Review of Performance Metrics for Computer-Aided Detection Systems

by Doohyun Park

Bioengineering 2024, 11(11), 1165; https://doi.org/10.3390/bioengineering11111165 - 19 Nov 2024

Viewed by 1285

Abstract

This paper aims to provide a structured analysis of the performance metrics used in computer-aided detection (CAD) systems, specifically focusing on lung nodule detection in computed tomography (CT) images. By examining key metrics along with their respective strengths and limitations, this study offers [...] Read more.

This paper aims to provide a structured analysis of the performance metrics used in computer-aided detection (CAD) systems, specifically focusing on lung nodule detection in computed tomography (CT) images. By examining key metrics along with their respective strengths and limitations, this study offers guidelines to assist in selecting appropriate metrics. Evaluation methods for CAD systems for lung nodule detection are primarily categorized into per-scan and per-nodule approaches. For per-scan analysis, a key metric is the area under the receiver operating characteristic (ROC) curve (AUROC), which evaluates the ability of the system to distinguish between scans with and without nodules. For per-nodule analysis, the nodule-level sensitivity at fixed false positives per scan is often used, supplemented by the free-response receiver operating characteristic (FROC) curve and the competition performance metric (CPM). However, the CPM does not provide normalized scores because it theoretically ranges from zero to infinity and largely varies depending on the characteristics of the data. To address the advantages and limitations of ROC and FROC curves, an alternative FROC (AFROC) was introduced to combine the strengths of both per-scan and per-nodule analyses. This paper discusses the principles of each metric and their relative strengths, providing insights into their clinical implications and practical utility. Full article

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

► Show Figures

Figure 1

29 pages, 1081 KiB

Open AccessReview

Hydrogel-Enhanced Autologous Chondrocyte Implantation for Cartilage Regeneration—An Update on Preclinical Studies

by Xenab Ahmadpoor, Jessie Sun, Nerone Douglas, Weimin Zhu and Hang Lin

Bioengineering 2024, 11(11), 1164; https://doi.org/10.3390/bioengineering11111164 - 19 Nov 2024

Cited by 1 | Viewed by 1607

Abstract

Autologous chondrocyte implantation (ACI) and matrix-induced ACI (MACI) have demonstrated improved clinical outcomes and reduced revision rates for treating osteochondral and chondral defects. However, their ability to achieve lasting, fully functional repair remains limited. To overcome these challenges, scaffold-enhanced ACI, particularly utilizing hydrogel-based [...] Read more.

Autologous chondrocyte implantation (ACI) and matrix-induced ACI (MACI) have demonstrated improved clinical outcomes and reduced revision rates for treating osteochondral and chondral defects. However, their ability to achieve lasting, fully functional repair remains limited. To overcome these challenges, scaffold-enhanced ACI, particularly utilizing hydrogel-based biomaterials, has emerged as an innovative strategy. These biomaterials are intended to mimic the biological composition, structural organization, and biomechanical properties of native articular cartilage. This review aims to provide comprehensive and up-to-date information on advancements in hydrogel-enhanced ACI from the past decade. We begin with a brief introduction to cartilage biology, mechanisms of cartilage injury, and the evolution of surgical techniques, particularly looking at ACI. Subsequently, we review the diversity of hydrogel scaffolds currently undergoing development and evaluation in preclinical studies for articular cartilage regeneration, emphasizing chondrocyte-laden hydrogels applicable to ACI. Finally, we address the key challenges impeding effective clinical translation, with particular attention to issues surrounding fixation and integration, aiming to inform and guide the future progression of tissue engineering strategies. Full article

(This article belongs to the Section Regenerative Engineering)

► Show Figures

Figure 1

30 pages, 8578 KiB

Open AccessEditor’s ChoiceArticle

Around-Body Versus On-Body Motion Sensing: A Comparison of Efficacy Across a Range of Body Movements and Scales

by Katelyn Rohrer, Luis De Anda, Camila Grubb, Zachary Hansen, Jordan Rodriguez, Greyson St Pierre, Sara Sheikhlary, Suleyman Omer, Binh Tran, Mehrail Lawendy, Farah Alqaraghuli, Chris Hedgecoke, Youssif Abdelkeder, Rebecca C. Slepian, Ethan Ross, Ryan Chung and Marvin J. Slepian

Bioengineering 2024, 11(11), 1163; https://doi.org/10.3390/bioengineering11111163 - 19 Nov 2024

Viewed by 1079

Abstract

Motion is vital for life. Currently, the clinical assessment of motion abnormalities is largely qualitative. We previously developed methods to quantitatively assess motion using visual detection systems (around-body) and stretchable electronic sensors (on-body). Here we compare the efficacy of these methods across predefined [...] Read more.

Motion is vital for life. Currently, the clinical assessment of motion abnormalities is largely qualitative. We previously developed methods to quantitatively assess motion using visual detection systems (around-body) and stretchable electronic sensors (on-body). Here we compare the efficacy of these methods across predefined motions, hypothesizing that the around-body system detects motion with similar accuracy as on-body sensors. Six human volunteers performed six defined motions covering three excursion lengths, small, medium, and large, which were analyzed via both around-body visual marker detection (MoCa version 1.0) and on-body stretchable electronic sensors (BioStamp version 1.0). Data from each system was compared as to the extent of trackability and comparative efficacy between systems. Both systems successfully detected motions, allowing quantitative analysis. Angular displacement between systems had the highest agreement efficiency for the bicep curl and body lean motion, with 73.24% and 65.35%, respectively. The finger pinch motion had an agreement efficiency of 36.71% and chest abduction/adduction had 45.55%. Shoulder abduction/adduction and shoulder flexion/extension motions had the lowest agreement efficiencies with 24.49% and 26.28%, respectively. MoCa was comparable to BioStamp in terms of angular displacement, though velocity and linear speed output could benefit from additional processing. Our findings demonstrate comparable efficacy for non-contact motion detection to that of on-body sensor detection, and offers insight as to the best system selection for specific clinical uses based on the use-case of the desired motion being analyzed. Full article

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

► Show Figures

Figure 1

20 pages, 5423 KiB

Open AccessArticle

Intelligent Evaluation Method for Scoliosis at Home Using Back Photos Captured by Mobile Phones

by Yongsheng Li, Xiangwei Peng, Qingyou Mao, Mingjia Ma, Jiaqi Huang, Shuo Zhang, Shaojie Dong, Zhihui Zhou, Yue Lan, Yu Pan, Ruimou Xie, Peiwu Qin and Kehong Yuan

Bioengineering 2024, 11(11), 1162; https://doi.org/10.3390/bioengineering11111162 - 18 Nov 2024

Cited by 1 | Viewed by 1451

Abstract

The traditional scoliosis examination based on X-ray film is not suitable for large-scale screening, and it is also not suitable for dynamic evaluation during rehabilitation. Therefore, based on computer vision technology, this paper puts forward an evaluation method of scoliosis with different photos [...] Read more.

The traditional scoliosis examination based on X-ray film is not suitable for large-scale screening, and it is also not suitable for dynamic evaluation during rehabilitation. Therefore, based on computer vision technology, this paper puts forward an evaluation method of scoliosis with different photos of the back taken by mobile phones, which involves three aspects: first, based on the key point detection model of YOLOv8, an algorithm for judging the type of spinal coronal curvature is proposed; second, an algorithm for evaluating the coronal plane of the spine based on the key points of the human back is proposed, aiming at quantifying the deviation degree of the spine in the coronal plane; third, the measurement algorithm of trunk rotation (ATR angle) based on multi-scale automatic peak detection (AMPD) is proposed, aiming at quantifying the deviation degree of the spine in sagittal plane. The public dataset and clinical paired data (mobile phone photo and X-ray) are used to test. The results show that this method has high accuracy and effectiveness in distinguishing the type of spinal curvature and evaluating the degree of deviation, which is higher than other deep learning algorithms. Full article

(This article belongs to the Special Issue New Sight of Intelligent Algorithm Model and Medical Device in Bioengineering: Updates and Direction)

► Show Figures

Graphical abstract

13 pages, 5606 KiB

Open AccessArticle

Influence of Bone Conditions on the Accuracy of Implant Placement

by Zhicheng Gong, Yuyin Shen, Shengcai Qi, Lai Cao, Xinyi Fan, Chunhui Lu and Jue Wang

Bioengineering 2024, 11(11), 1161; https://doi.org/10.3390/bioengineering11111161 - 18 Nov 2024

Cited by 2 | Viewed by 1265

Abstract

This study aimed to assess the influence of cortical bone thickness, bone density, and residual ridge morphology in the posterior mandibular area on the accuracy of implant placement using tooth-supported digital guides. The research included 75 implants from 55 patients. Each patient underwent [...] Read more.

This study aimed to assess the influence of cortical bone thickness, bone density, and residual ridge morphology in the posterior mandibular area on the accuracy of implant placement using tooth-supported digital guides. The research included 75 implants from 55 patients. Each patient underwent a cone-beam computed tomography (CBCT) scan for image analysis. Simplant^® Pro 17 software (SIMPLANT Pro 17.01) was utilized to measure cortical bone thickness, bone density, and residual ridge morphology at the implant sites. Subsequently, 3Shape Trios software (3Shape TRIOS Design Studio 1.7.19.0) was applied to delineate optimal implant positions and design tooth-supported surgical guides. After implant treatment, the linear and angular deviations from the planned placement were quantified. Multiple linear regression, Kruskal–Wallis test, Conover–Iman test, and Bonferroni adjustment were conducted to investigate the impact of bone characteristics on implant placement precision. The tooth-supported digital guides used in this study were sufficient to fulfill the precision criteria for implant treatment. Bone density was found to significantly influence the buccal-lingual angular deviation, mesio-distal linear deviation, and mesio-distal angular deviation (p < 0.05). Additionally, significant variances were noted in the coronal deviation, apical deviation and depth deviation in buccal-lingual orientation, coronal deviation, and apical deviation in mesio-distal orientation across various residual ridge morphologies (p < 0.05). Low bone density and S-shape morphology may affect the accuracy of implant placement using tooth-supported surgical guides. Full article

(This article belongs to the Special Issue Mechanobiology in Biomedical Engineering)

► Show Figures

Figure 1

25 pages, 1793 KiB

Open AccessReview

Mesenchymal Stem Cells and Their Extracellular Vesicles Are a Promising Alternative to Antibiotics for Treating Sepsis

by Yu Jiang, Yunjuan Song, Qin Zeng and Bin Jiang

Bioengineering 2024, 11(11), 1160; https://doi.org/10.3390/bioengineering11111160 - 18 Nov 2024

Cited by 1 | Viewed by 2322

Abstract

Sepsis is a life-threatening disease caused by the overwhelming response to pathogen infections. Currently, treatment options for sepsis are limited to broad-spectrum antibiotics and supportive care. However, the growing resistance of pathogens to common antibiotics complicates treatment efforts. Excessive immune response (i.e., cytokine [...] Read more.

Sepsis is a life-threatening disease caused by the overwhelming response to pathogen infections. Currently, treatment options for sepsis are limited to broad-spectrum antibiotics and supportive care. However, the growing resistance of pathogens to common antibiotics complicates treatment efforts. Excessive immune response (i.e., cytokine storm) can persist even after the infection is cleared. This overactive inflammatory response can severely damage multiple organ systems. Given these challenges, managing the excessive immune response is critical in controlling sepsis progression. Therefore, Mesenchymal stem cells (MSCs), with their immunomodulatory and antibacterial properties, have emerged as a promising option for adjunctive therapy in treating sepsis. Moreover, MSCs exhibit a favorable safety profile, as they are eventually eliminated by the host’s immune system within several months post-administration, resulting in minimal side effects and have not been linked to common antibiotic therapy drawbacks (i.e., antibiotic resistance). This review explores the potential of MSCs as a personalized therapy for sepsis treatment, clarifying their mechanisms of action and providing up-to-date technological advancements to enhance their protective efficacy for patients suffering from sepsis and its consequences. Full article

(This article belongs to the Special Issue Innovations in Regenerative Therapy: Cell and Cell-Free Approaches)

► Show Figures

Graphical abstract

9 pages, 2627 KiB

Open AccessArticle

Artificial Intelligence Diagnosing of Oral Lichen Planus: A Comparative Study

by Sensen Yu, Wansu Sun, Dawei Mi, Siyu Jin, Xing Wu, Baojian Xin, Hengguo Zhang, Yuanyin Wang, Xiaoyu Sun and Xin He

Bioengineering 2024, 11(11), 1159; https://doi.org/10.3390/bioengineering11111159 - 18 Nov 2024

Cited by 4 | Viewed by 1600

Abstract

Early diagnosis of oral lichen planus (OLP) is challenging, which traditionally is dependent on clinical experience and subjective interpretation. Artificial intelligence (AI) technology has been widely applied in objective and rapid diagnoses. In this study, we aim to investigate the potential of AI [...] Read more.

Early diagnosis of oral lichen planus (OLP) is challenging, which traditionally is dependent on clinical experience and subjective interpretation. Artificial intelligence (AI) technology has been widely applied in objective and rapid diagnoses. In this study, we aim to investigate the potential of AI diagnosis in OLP and evaluate its effectiveness in improving diagnostic accuracy and accelerating clinical decision making. A total of 128 confirmed OLP patients were included, and lesion images from various anatomical sites were collected. The diagnosis was performed using AI platforms, including ChatGPT-4O, ChatGPT (Diagram-Date extension), and Claude Opus, for AI directly identification and AI pre-training identification. After OLP feature training, the diagnostic accuracy of the AI platforms significantly improved, with the overall recognition rates of ChatGPT-4O, ChatGPT (Diagram-Date extension), and Claude Opus increasing from 59%, 68%, and 15% to 77%, 80%, and 50%, respectively. Additionally, the pre-training recognition rates for buccal mucosa reached 94%, 93%, and 56%, respectively. However, the AI platforms performed less effectively when recognizing lesions in less common sites and complex cases; for instance, the pre-training recognition rates for the gums were only 60%, 60%, and 20%, demonstrating significant limitations. The study highlights the strengths and limitations of different AI technologies and provides a reference for future AI applications in oral medicine. Full article

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

► Show Figures

Graphical abstract

16 pages, 4966 KiB

Open AccessArticle

Developing a Virtual Model of the Rhesus Macaque Inner Ear

by Cayman Matson, Nicholas Castle and Chenkai Dai

Bioengineering 2024, 11(11), 1158; https://doi.org/10.3390/bioengineering11111158 - 18 Nov 2024

Viewed by 967

Abstract

A virtual model of the rhesus macaque inner ear was created in the present study. Rhesus macaques have been valuable in cochlear research; however, their high cost prompts a need for alternative methods. Finite Element (FE) analysis offers a promising solution by enabling [...] Read more.

A virtual model of the rhesus macaque inner ear was created in the present study. Rhesus macaques have been valuable in cochlear research; however, their high cost prompts a need for alternative methods. Finite Element (FE) analysis offers a promising solution by enabling detailed simulations of the inner ear. This study employs FE analysis to create a virtual model of the rhesus macaque’s inner ear, reconstructed from MRI scans, to explore how cochlear implants (CIs) impact residual hearing loss. Harmonic-acoustic simulations of sound wave transmission indicate that CIs have minor effects on the displacement of the basilar membrane and thus minimally impact residual hearing loss post-implantation, but stiffening of the round window membrane worsens this effect. While the rhesus macaque FE model presented in this study shows some promise, its potential applications will require further validation through additional simulations and experimental studies. Full article

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

► Show Figures

Figure 1

14 pages, 7048 KiB

Open AccessArticle

Classification of Dog Breeds Using Convolutional Neural Network Models and Support Vector Machine

by Ying Cui, Bixia Tang, Gangao Wu, Lun Li, Xin Zhang, Zhenglin Du and Wenming Zhao

Bioengineering 2024, 11(11), 1157; https://doi.org/10.3390/bioengineering11111157 - 17 Nov 2024

Cited by 1 | Viewed by 2452

Abstract

When classifying breeds of dogs, the accuracy of classification significantly affects breed identification and dog research. Using images to classify dog breeds can improve classification efficiency; however, it is increasingly challenging due to the diversities and similarities among dog breeds. Traditional image classification [...] Read more.

When classifying breeds of dogs, the accuracy of classification significantly affects breed identification and dog research. Using images to classify dog breeds can improve classification efficiency; however, it is increasingly challenging due to the diversities and similarities among dog breeds. Traditional image classification methods primarily rely on extracting simple geometric features, while current convolutional neural networks (CNNs) are capable of learning high-level semantic features. However, the diversity of dog breeds continues to pose a challenge to classification accuracy. To address this, we developed a model that integrates multiple CNNs with a machine learning method, significantly improving the accuracy of dog images classification. We used the Stanford Dog Dataset, combined image features from four CNN models, filtered the features using principal component analysis (PCA) and gray wolf optimization algorithm (GWO), and then classified the features with support vector machine (SVM). The classification accuracy rate reached 95.24% for 120 breeds and 99.34% for 76 selected breeds, respectively, demonstrating a significant improvement over existing methods using the same Stanford Dog Dataset. It is expected that our proposed method will further serve as a fundamental framework for the accurate classification of a wider range of species. Full article

(This article belongs to the Section Biosignal Processing)

► Show Figures

Graphical abstract

11 pages, 1076 KiB

Open AccessArticle

Influence of Dextran Solution and Corneal Collagen Crosslinking on Corneal Biomechanical Parameters Evaluated by Corvis ST In Vitro

by Xiao Qin, Bi Hu, Lili Guo, Haixia Zhang, Lin Li, Ying Jie and Lei Tian

Bioengineering 2024, 11(11), 1156; https://doi.org/10.3390/bioengineering11111156 - 17 Nov 2024

Viewed by 979

Abstract

Purpose: To analyze the influence of dextran solution and corneal collagen crosslinking (CXL) on corneal biomechanical parameters in vitro, evaluated by Corneal Visualization Scheimpflug Technology (Corvis ST). Materials and Methods: Forty porcine eyes were included in this study. Twenty porcine eyes were instilled [...] Read more.

Purpose: To analyze the influence of dextran solution and corneal collagen crosslinking (CXL) on corneal biomechanical parameters in vitro, evaluated by Corneal Visualization Scheimpflug Technology (Corvis ST). Materials and Methods: Forty porcine eyes were included in this study. Twenty porcine eyes were instilled with dextran solution for 30 min (10 eyes in 2% dextran solution and 10 eyes in 20% dextran solution). CXL treatment was performed in 10 porcine eyes; the other 10 porcine eyes were regarded as the control group. Each eye was fixed on an experimental inflation platform to carry out Corvis measurements at different IOPs. Corneal biomechanical parameters were calculated based on Corvis measurement. Statistical analysis was used to analyze the influence of dextran solution and CXL on corneal biomechanical parameters based on Corvis parameters. Results: The corneal energy-absorbed area (A_absorbed) decreased after being instilled with dextran solution under IOP of 15 mmHg (p < 0.001); the elastic modulus (E) of the cornea instilled with 20% dextran solution was significantly higher than that instilled with 2% dextran solution (p < 0.001), since it decreased after being instilled with 20% dextran solution (p = 0.030); the stiffness parameter at the first applanation (SP-A1) increased after CXL (p < 0.001). Conclusions: Both dextran solution and CXL can change corneal biomechanical properties; the concentration of dextran solution can influence the corneal biomechanical properties, which may, in turn, affect the effectiveness of CXL. SP-A1 may be used as an effective parameter for the evaluation of CXL. Full article

(This article belongs to the Special Issue Biomechanics Studies in Ophthalmology)

► Show Figures

Figure 1

13 pages, 4744 KiB

Open AccessArticle

Preclinical Experimental Study on New Cervical Implant Design to Improve Peri-Implant Tissue Healing

by Sergio Alexandre Gehrke, Guillermo Castro Cortellari, Jaime Aramburú Júnior, Tiago Luis Eilers Treichel, Marco Aurelio Bianchini, Antonio Scarano and Piedad N. De Aza

Bioengineering 2024, 11(11), 1155; https://doi.org/10.3390/bioengineering11111155 - 16 Nov 2024

Viewed by 1368

Abstract

Objectives: In this preclinical study, we used an experimental rabbit model to investigate the effects of a new implant design that involves specific changes to the cervical portion, using a conventional implant design in the control group. Materials and Methods: We used 10 [...] Read more.

Objectives: In this preclinical study, we used an experimental rabbit model to investigate the effects of a new implant design that involves specific changes to the cervical portion, using a conventional implant design in the control group. Materials and Methods: We used 10 rabbits and 40 dental implants with two different macrogeometries. Two groups were formed (n = 20 per group): the Collo group, wherein implants with the new cervical design were used, which present a concavity (reduction in diameter) in the first 3.5 mm, the portion without surface treatment; the Control group, wherein conical implants with the conventional design were used, with surface treatment throughout the body. All implants were 4 mm in diameter and 10 mm in length. The initial implant stability quotient (ISQ) was measured immediately after the implant insertion (T1) and sample removal (T2 and T3). The animals (n = five animals/time) were euthanized at 3 weeks (T1) and 4 weeks (T2). Histological sections were prepared and the bone–implant contact (BIC%) and tissue area fraction occupancy (TAFO%) percentages were analyzed in the predetermined cervical area; namely, the first 4 mm from the implant platform. Results: The ISQ values showed no statistical differences at T1 and T2 (p = 0.9458 and p = 0.1103, respectively) between the groups. However, at T3, higher values were found for the Collo group (p = 0.0475) than those found for the Control group. The Collo samples presented higher BIC% values than those of the Control group, with statistical differences of p = 0.0009 at 3 weeks and p = 0.0007 at 4 weeks. There were statistical differences in the TAFO% (new bone, medullary spaces, and the collagen matrix) between the groups at each evaluation time (p < 0.001). Conclusions: Considering the limitations of the present preclinical study, the results demonstrate that the new implant design (the Collo group) had higher implant stability (ISQ) values in the samples after 4 weeks of implantation. Furthermore, the histomorphometric BIC% and TAFO% analyses showed that the Collo group had higher values at both measurement times than the Control group did. These findings indicate that changes made to the cervical design of the Collo group implants may benefit the maintenance of peri-implant tissue health. Full article

► Show Figures

Figure 1

15 pages, 1081 KiB

Open AccessReview

Introduction of AI Technology for Objective Physical Function Assessment

by Nobuji Kouno, Satoshi Takahashi, Masaaki Komatsu, Yusuke Sakaguchi, Naoaki Ishiguro, Katsuji Takeda, Kyoko Fujioka, Ayumu Matsuoka, Maiko Fujimori and Ryuji Hamamoto

Bioengineering 2024, 11(11), 1154; https://doi.org/10.3390/bioengineering11111154 - 16 Nov 2024

Cited by 1 | Viewed by 1225

Abstract

Objective physical function assessment is crucial for determining patient eligibility for treatment and adjusting the treatment intensity. Existing assessments, such as performance status, are not well standardized, despite their frequent use in daily clinical practice. This paper explored how artificial intelligence (AI) could [...] Read more.

Objective physical function assessment is crucial for determining patient eligibility for treatment and adjusting the treatment intensity. Existing assessments, such as performance status, are not well standardized, despite their frequent use in daily clinical practice. This paper explored how artificial intelligence (AI) could predict physical function scores from various patient data sources and reviewed methods to measure objective physical function using this technology. This review included relevant articles published in English that were retrieved from PubMed. These studies utilized AI technology to predict physical function indices from patient data extracted from videos, sensors, or electronic health records, thereby eliminating manual measurements. Studies that used AI technology solely to automate traditional evaluations were excluded. These technologies are recommended for future clinical systems that perform repeated objective physical function assessments in all patients without requiring extra time, personnel, or resources. This enables the detection of minimal changes in a patient’s condition, enabling early intervention and enhanced outcomes. Full article

(This article belongs to the Special Issue ML and AI for Augmented Biosensing Applications)

► Show Figures

Figure 1

20 pages, 2362 KiB

Open AccessArticle

Machine Learning-Driven GLCM Analysis of Structural MRI for Alzheimer’s Disease Diagnosis

by Maria João Oliveira, Pedro Ribeiro and Pedro Miguel Rodrigues

Bioengineering 2024, 11(11), 1153; https://doi.org/10.3390/bioengineering11111153 - 15 Nov 2024

Cited by 1 | Viewed by 1693

Abstract

Background: Alzheimer’s disease (AD) is a progressive and irreversible neurodegenerative condition that increasingly impairs cognitive functions and daily activities. Given the incurable nature of AD and its profound impact on the elderly, early diagnosis (at the mild cognitive impairment (MCI) stage) and intervention [...] Read more.

Background: Alzheimer’s disease (AD) is a progressive and irreversible neurodegenerative condition that increasingly impairs cognitive functions and daily activities. Given the incurable nature of AD and its profound impact on the elderly, early diagnosis (at the mild cognitive impairment (MCI) stage) and intervention are crucial, focusing on delaying disease progression and improving patients’ quality of life. Methods: This work aimed to develop an automatic sMRI-based method to detect AD in three different stages, namely healthy controls (CN), mild cognitive impairment (MCI), and AD itself. For such a purpose, brain sMRI images from the ADNI database were pre-processed, and a set of 22 texture statistical features from the sMRI gray-level co-occurrence matrix (GLCM) were extracted from various slices within different anatomical planes. Different combinations of features and planes were used to feed classical machine learning (cML) algorithms to analyze their discrimination power between the groups. Results: The cML algorithms achieved the following classification accuracy: 85.2% for AD vs. CN, 98.5% for AD vs. MCI, 95.1% for CN vs. MCI, and 87.1% for all vs. all. Conclusions: For the pair AD vs. MCI, the proposed model outperformed state-of-the-art imaging source studies by 0.1% and non-imaging source studies by 4.6%. These results are particularly significant in the field of AD classification, opening the door to more efficient early diagnosis in real-world settings since MCI is considered a precursor to AD. Full article

(This article belongs to the Special Issue Cutting-Edge Applications of Machine and Deep Learning in Biomedical Signal and Image Processing)

► Show Figures

Figure 1

21 pages, 5275 KiB

Open AccessArticle

Classical Batch Distillation of Anaerobic Digestate to Isolate Ammonium Bicarbonate: Membrane Not Necessary!

by Alejandro Moure Abelenda and Jonas Baltrusaitis

Bioengineering 2024, 11(11), 1152; https://doi.org/10.3390/bioengineering11111152 - 15 Nov 2024

Cited by 1 | Viewed by 1147

Abstract

The excessive mineralization of organic molecules during anaerobic fermentation increases the availability of nitrogen and carbon. For this reason, the development of downstream processing technologies is required to better manage ammonia and carbon dioxide emissions during the storage and land application of the [...] Read more.

The excessive mineralization of organic molecules during anaerobic fermentation increases the availability of nitrogen and carbon. For this reason, the development of downstream processing technologies is required to better manage ammonia and carbon dioxide emissions during the storage and land application of the resulting soil organic amendment. The present work investigated classical distillation as a technology for valorizing ammoniacal nitrogen (NH₄⁺-N) in anaerobic digestate. The results implied that the direct isolation of ammonium bicarbonate (NH₄HCO₃) was possible when applying the reactive distillation to the food waste digestate (FWD) with a high content of NH₄⁺-N, while the addition of antifoam to the agrowaste digestate (AWD) was necessary to be able to produce an aqueous solution of NH₄HCO₃ as the distillate. The reason was that the extraction of NH₄HCO₃ from the AWD required a higher temperature (>95 °C) and duration (i.e., steady state in batch operation) than the recovery of the inorganic fertilizer from the FWD. The titration method, when applied to the depleted digestate, offered the quickest way of monitoring the reactive distillation because the buffer capacity of the distillate was much higher. The isolation of NH₄HCO₃ from the FWD was attained in a transient mode at a temperature below 90 °C (i.e., while heating up to reach the desired distillation temperature or cooling down once the batch distillation was finished). For the operating conditions to be regarded as techno-economically feasible, they should be attained in the anaerobic digestion plant by integrating the heat harvested from the engines, which convert the biogas into electricity. Full article

(This article belongs to the Special Issue From Residues to Bio-Based Products through Bioprocess Engineering)

► Show Figures

Graphical abstract

17 pages, 9733 KiB

Open AccessArticle

Raman Handheld Versus Microscopic Spectroscopy for Estimating the Post-Mortem Interval of Human Bones: A Comparative Pilot Study

by Johannes Dominikus Pallua, Christina Louis, Nicole Gattermair, Andrea Brunner, Bettina Zelger, Michael Schirmer, Jovan Badzoka, Christoph Kappacher, Christian Wolfgang Huck, Jürgen Popp, Walter Rabl and Claudia Wöss

Bioengineering 2024, 11(11), 1151; https://doi.org/10.3390/bioengineering11111151 - 15 Nov 2024

Viewed by 1404

Abstract

The post-mortem interval estimation for human skeletal remains is critical in forensic medicine. This study used Raman spectroscopy, specifically comparing a handheld device to a Raman microscope for PMI estimations. Analyzing 99 autopsy bone samples and 5 archeological samples, the research categorized them [...] Read more.

The post-mortem interval estimation for human skeletal remains is critical in forensic medicine. This study used Raman spectroscopy, specifically comparing a handheld device to a Raman microscope for PMI estimations. Analyzing 99 autopsy bone samples and 5 archeological samples, the research categorized them into five PMI classes using conventional methods. Key parameters—like ν₁PO₄³⁻ intensity and crystallinity—were measured and analyzed. A principal component analysis effectively distinguished between PMI classes, indicating high classification accuracy for both devices. While both methods proved reliable, the fluorescence interference presented challenges in accurately determining the age of archeological samples. Ultimately, the study highlighted how Raman spectroscopy could enhance PMI estimation accuracy, especially in non-specialized labs, suggesting the potential for improved device optimization in the field. Full article

(This article belongs to the Special Issue Optical Imaging for Biomedical Applications)

► Show Figures

Figure 1

15 pages, 2419 KiB

Open AccessArticle

Deciphering Membrane Proteins Through Deep Learning Models by Revealing Their Locale Within the Cell

by Mehwish Faiz, Saad Jawaid Khan, Fahad Azim, Nazia Ejaz and Fahad Shamim

Bioengineering 2024, 11(11), 1150; https://doi.org/10.3390/bioengineering11111150 - 15 Nov 2024

Viewed by 931

Abstract

Membrane proteins constitute essential biomolecules attached to or integrated into cellular and organelle membranes, playing diverse roles in cellular processes. Their precise localization is crucial for understanding their functions. Existing protein subcellular localization predictors are predominantly trained on globular proteins; their performance diminishes [...] Read more.

Membrane proteins constitute essential biomolecules attached to or integrated into cellular and organelle membranes, playing diverse roles in cellular processes. Their precise localization is crucial for understanding their functions. Existing protein subcellular localization predictors are predominantly trained on globular proteins; their performance diminishes for membrane proteins, explicitly via deep learning models. To address this challenge, the proposed study segregates membrane proteins into three distinct locations, including the plasma membrane, internal membrane, and membrane of the organelle, using deep learning algorithms including recurrent neural networks (RNN) and Long Short-Term Memory (LSTM). A redundancy-curtailed dataset of 3000 proteins from the MemLoci approach is selected for the investigation, along with incorporating pseudo amino acid composition (PseAAC). PseAAC is an exemplary technique for extracting protein information hidden in the amino acid sequences. After extensive testing, the results show that the accuracy for LSTM and RNN is 83.4% and 80.5%, respectively. The results show that the LSTM model outperforms the RNN and is most commonly employed in proteomics. Full article

(This article belongs to the Special Issue Bio-Macromolecular Modeling and Computational Design)

► Show Figures

Graphical abstract

19 pages, 11979 KiB

Open AccessArticle

Residual Stress Homogenization of Hybrid Implants

by Marta Sanjuán Álvarez, Daniel Robles, Javier Gil Mur, Saray Fernández-Hernández, Esteban Pérez-Pevida and Aritza Brizuela-Velasco

Bioengineering 2024, 11(11), 1149; https://doi.org/10.3390/bioengineering11111149 - 15 Nov 2024

Viewed by 945

Abstract

Objectives: Hybrid implants commonly exhibit decreased corrosion resistance and fatigue due to differences in compressive residual stresses between the smooth and rough surfaces. The main objective of this study was to investigate the influence of an annealing heat treatment to reduce the residual [...] Read more.

Objectives: Hybrid implants commonly exhibit decreased corrosion resistance and fatigue due to differences in compressive residual stresses between the smooth and rough surfaces. The main objective of this study was to investigate the influence of an annealing heat treatment to reduce the residual stresses in hybrid implants. Methodology: Commercially pure titanium (CpTi) bars were heat-treated at 800 °C and different annealing times. Optical microscopy was used to analyze the resulting grain growth kinetics. Diffractometry was used to measure residual stress after heat treatment, corrosion resistance by open circuit potential (E_OCP), corrosion potentials (E_CORR), and corrosion currents (I_CORR) of heat-treated samples, as well as fatigue behavior by creep testing. The von Mises distribution and the resulting microstrains in heat-treated hybrid implants and in cortical and trabecular bone were assessed by finite element analysis. The results of treated hybrid implants were compared to those of untreated hybrid implants and hybrid implants with a rough surface (shot-blasted). Results: The proposed heat treatment (800 °C for 30 min, followed by quenching in water at 20 °C) could successfully homogenize the residual stress difference between the two surfaces of the hybrid implant (−20.2 MPa). It provides better fatigue behavior and corrosion resistance (p ˂ 0.05, ANOVA). Stress distribution was significantly improved in the trabecular bone. Heat-treated hybrid implants performed worse than implants with a rough surface. Clinical significance: Annealing heat treatment can be used to improve the mechanical properties and corrosion resistance of hybrid surface implants by homogenizing residual stresses. Full article

(This article belongs to the Special Issue Application of Bioengineering to Dentistry)

► Show Figures

Graphical abstract

19 pages, 1364 KiB

Open AccessArticle

A New Breast Cancer Discovery Strategy: A Combined Outlier Rejection Technique and an Ensemble Classification Method

by Shereen H. Ali and Mohamed Shehata

Bioengineering 2024, 11(11), 1148; https://doi.org/10.3390/bioengineering11111148 - 15 Nov 2024

Cited by 1 | Viewed by 988

Abstract

Annually, many people worldwide lose their lives due to breast cancer, making it one of the most prevalent cancers in the world. Since the disease is becoming more common, early detection of breast cancer is essential to avoiding serious complications and possibly death [...] Read more.

Annually, many people worldwide lose their lives due to breast cancer, making it one of the most prevalent cancers in the world. Since the disease is becoming more common, early detection of breast cancer is essential to avoiding serious complications and possibly death as well. This research provides a novel Breast Cancer Discovery (BCD) strategy to aid patients by providing prompt and sensitive detection of breast cancer. The two primary steps that form the BCD are the Breast Cancer Discovery Step (BCDS) and the Pre-processing Step (P²S). In the P²S, the needed data are filtered from any non-informative data using three primary operations: data normalization, feature selection, and outlier rejection. Only then does the diagnostic model in the BCDS for precise diagnosis begin to be trained. The primary contribution of this research is the novel outlier rejection technique known as the Combined Outlier Rejection Technique (CORT). CORT is divided into two primary phases: (i) the Quick Rejection Phase (QRP), which is a quick phase utilizing a statistical method, and (ii) the Accurate Rejection Phase (ARP), which is a precise phase using an optimization method. Outliers are rapidly eliminated during the QRP using the standard deviation, and the remaining outliers are thoroughly eliminated during ARP via Binary Harris Hawk Optimization (BHHO). The P²S in the BCD strategy indicates that data normalization is a pre-processing approach used to find numeric values in the datasets that fall into a predetermined range. Information Gain (IG) is then used to choose the optimal subset of features, and CORT is used to reject incorrect training data. Furthermore, based on the filtered data from the P²S, an Ensemble Classification Method (ECM) is utilized in the BCDS to identify breast cancer patients. This method consists of three classifiers: Naïve Bayes (NB), K-Nearest Neighbors (KNN), and Support Vector Machine (SVM). The Wisconsin Breast Cancer Database (WBCD) dataset, which contains digital images of fine-needle aspiration samples collected from patients’ breast masses, is used herein to compare the BCD strategy against several contemporary strategies. According to the outcomes of the experiment, the suggested method is very competitive. It achieves 0.987 accuracy, 0.013 error, 0.98 recall, 0.984 precision, and a run time of 3 s, outperforming all other methods from the literature. Full article

(This article belongs to the Special Issue Artificial Intelligence for Better Healthcare and Precision Medicine)

► Show Figures

Graphical abstract

14 pages, 6553 KiB

Open AccessArticle

An Arteriovenous Bioreactor Perfusion System for Physiological In Vitro Culture of Complex Vascularized Tissue Constructs

by Florian Helms, Delia Käding, Thomas Aper, Arjang Ruhparwar and Mathias Wilhelmi

Bioengineering 2024, 11(11), 1147; https://doi.org/10.3390/bioengineering11111147 - 14 Nov 2024

Viewed by 1008

Abstract

Background: The generation and perfusion of complex vascularized tissues in vitro requires sophisticated perfusion techniques. For multiscale arteriovenous networks, not only the arterial, but also the venous, biomechanical and biochemical conditions that physiologically exist in the human body must be accurately emulated. For [...] Read more.

Background: The generation and perfusion of complex vascularized tissues in vitro requires sophisticated perfusion techniques. For multiscale arteriovenous networks, not only the arterial, but also the venous, biomechanical and biochemical conditions that physiologically exist in the human body must be accurately emulated. For this, we here present a modular arteriovenous perfusion system for the in vitro culture of a multi-scale bioartificial vascular network. Methods: The custom-built perfusion system consisted of two circuits: in the arterial circuit, physiological arterial biomechanical and biochemical conditions were simulated using a modular set-up with a pulsatile peristaltic pump, compliance chambers, and resistors. In the venous circuit, venous conditions were emulated accordingly. In the center of the system, a bioartificial multi-scale vascularized fibrin-based tissue was perfused by both circuits simultaneously under biomimetic arteriovenous conditions. Culture conditions were monitored continuously using a multi-sensor monitoring system. Results: The physiological arterial and venous pressure- and flow-curves, as well as the microvascular arteriovenous oxygen partial pressure gradient, were accurately emulated in the perfusion system. The multi-sensor monitoring system facilitated live monitoring of the respective parameters and data-logging. In a proof-of-concept experiment, vascularized three-dimensional fibrin tissues showed sustained cell viability and homogenous microvessel formation after culture in the perfusion system. Conclusions: The arteriovenous perfusion system facilitated the in vitro culture of a multiscale vascularized tissue under physiological pressure-, flow-, and oxygen-gradient conditions. With that, it presents a promising technique for the in vitro generation and culture of complex large-scale vascularized tissues. Full article

(This article belongs to the Special Issue Bioreactors for Advanced Cell Culture, (Nano)toxicity, Regenerative Medicine and Organ Bioengineering—2nd Edition)

► Show Figures

Graphical abstract

25 pages, 4283 KiB

Open AccessArticle

Shape-Aware Adversarial Learning for Scribble-Supervised Medical Image Segmentation with a MaskMix Siamese Network: A Case Study of Cardiac MRI Segmentation

by Chen Li, Zhong Zheng and Di Wu

Bioengineering 2024, 11(11), 1146; https://doi.org/10.3390/bioengineering11111146 - 13 Nov 2024

Viewed by 1111

Abstract

The transition in medical image segmentation from fine-grained to coarse-grained annotation methods, notably scribble annotation, offers a practical and efficient preparation for deep learning applications. However, these methods often compromise segmentation precision and result in irregular contours. This study targets the enhancement of [...] Read more.

The transition in medical image segmentation from fine-grained to coarse-grained annotation methods, notably scribble annotation, offers a practical and efficient preparation for deep learning applications. However, these methods often compromise segmentation precision and result in irregular contours. This study targets the enhancement of scribble-supervised segmentation to match the accuracy of fine-grained annotation. Capitalizing on the consistency of target shapes across unpaired datasets, this study introduces a shape-aware scribble-supervised learning framework (MaskMixAdv) addressing two critical tasks: (1) Pseudo label generation, where a mixup-based masking strategy enables image-level and feature-level data augmentation to enrich coarse-grained scribbles annotations. A dual-branch siamese network is proposed to generate fine-grained pseudo labels. (2) Pseudo label optimization, where a CNN-based discriminator is proposed to refine pseudo label contours by distinguishing them from external unpaired masks during model fine-tuning. MaskMixAdv works under constrained annotation conditions as a label-efficient learning approach for medical image segmentation. A case study on public cardiac MRI datasets demonstrated that the proposed MaskMixAdv outperformed the state-of-the-art methods and narrowed the performance gap between scribble-supervised and mask-supervised segmentation. This innovation cuts annotation time by at least 95%, with only a minor impact on Dice performance, specifically a 2.6% reduction. The experimental outcomes indicate that employing efficient and cost-effective scribble annotation can achieve high segmentation accuracy, significantly reducing the typical requirement for fine-grained annotations. Full article

(This article belongs to the Section Biosignal Processing)

► Show Figures

Graphical abstract

15 pages, 3242 KiB

Open AccessReview

Unlocking the Potentials of Large Language Models in Orthodontics: A Scoping Review

by Jie Zheng, Xiaoqian Ding, Jingya Jane Pu, Sze Man Chung, Qi Yong H. Ai, Kuo Feng Hung and Zhiyi Shan

Bioengineering 2024, 11(11), 1145; https://doi.org/10.3390/bioengineering11111145 - 13 Nov 2024

Viewed by 1658

Abstract

(1) Background: In recent years, large language models (LLMs) such as ChatGPT have gained significant attention in various fields, including dentistry. This scoping review aims to examine the current applications and explore potential uses of LLMs in the orthodontic domain, shedding light on [...] Read more.

(1) Background: In recent years, large language models (LLMs) such as ChatGPT have gained significant attention in various fields, including dentistry. This scoping review aims to examine the current applications and explore potential uses of LLMs in the orthodontic domain, shedding light on how they might improve dental healthcare. (2) Methods: We carried out a comprehensive search in five electronic databases, namely PubMed, Scopus, Embase, ProQuest and Web of Science. Two authors independently screened articles and performed data extraction according to the eligibility criteria, following the PRISMA-ScR guideline. The main findings from the included articles were synthesized and analyzed in a narrative way. (3) Results: A total of 706 articles were searched, and 12 papers were eventually included. The applications of LLMs include improving diagnostic and treatment efficiency in orthodontics as well as enhancing communication with patients. (4) Conclusions: There is emerging research in countries worldwide on the use of LLMs in orthodontics, suggesting an upward trend in their acceptance within this field. However, the potential application of LLMs remains in its early stage, with a noticeable lack of extensive studies and tailored products to address specific clinical needs. Full article

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

► Show Figures

Graphical abstract

14 pages, 2756 KiB

Open AccessArticle

SCOUT: Skull-Corrected Optimization for Ultrasound Transducers

by Zheng Jiang, Michelle Hua, Jacqueline Li, Hieu Le Mau, James Choi, William B. Gormley, Jose M. Amich and Raahil M. Sha

Bioengineering 2024, 11(11), 1144; https://doi.org/10.3390/bioengineering11111144 - 13 Nov 2024

Viewed by 1855

Abstract

Transcranial focused ultrasound has been studied for non-invasive and localized treatment of many brain diseases. The biggest challenge for focusing ultrasound onto the brain is the skull, which attenuates ultrasound and changes its propagation direction, leading to pressure drop, focus shift, and defocusing. [...] Read more.

Transcranial focused ultrasound has been studied for non-invasive and localized treatment of many brain diseases. The biggest challenge for focusing ultrasound onto the brain is the skull, which attenuates ultrasound and changes its propagation direction, leading to pressure drop, focus shift, and defocusing. We presented an optimization algorithm which automatically found the optimal location for placing a single-element focused transducer. At this optimal location, the focus shift was in an acceptable range and the ultrasound was tightly focused. The algorithm simulated the beam profiles of placing the transducer at different locations and compared the results. Locations with a normalized peak-negative pressure (PNP) above threshold were first found. Then, the optimal location was identified as the location with the smallest focal volume. The optimal location found in this study had a normalized PNP of 0.966 and a focal volume of 6.8% smaller than without the skull. A Zeta navigation system was used to automatically place the transducer and track the error caused by movement. These results demonstrated that the algorithm could find the optimal transducer location to avoid large focus shift and defocusing. With the Zeta navigation system, our algorithm can help to make transcranial focused ultrasound treatment safer and more successful. Full article

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

► Show Figures

Figure 1

Journal Menu

Journal Browser

Bioengineering, Volume 11, Issue 11 (November 2024) – 123 articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI