Biomedical Engineering and Computational Biology最新文献

Background: Telemedicine facilitates remote consultations and expands access to healthcare, marking a transformative shift in the medical field. Given the critical role of surgeons in the healthcare system, the adoption of telemedicine in surgical practice offers both distinct benefits and challenges. This research aims to assess the predictors of telemedicine attitudes and usage among surgeons in Jeddah, Saudi Arabia.

Methods: An analytical cross-sectional study was carried out among 198 surgeons from public and private hospitals in Jeddah using convenience sampling technique. Data were collected in person using a pre-designed and validated questionnaire. Data analysis was carried out by IBM SPSS version 26. Chi-square tests and binary logistic regression were used to identify significant factors influencing surgeons' attitudes and usage of telemedicine.

Results: Among the participants, 54.5% reported having used telemedicine at least once in the past. Bivariate analysis revealed that surgeons in private hospitals (64.9%) were more likely to use telemedicine than those in public hospitals (40.4%; P = .001). Females were also associated with a higher usage (67.5%) in comparison to males (45.7%; P = .003). Frequent users were found to have less positive attitude compared to occasional users (35.4% vs 60.7%) (P < .001). Key concerns shaping attitudes toward telemedicine included limited ability to perform physical examinations, with 32.8% strongly agreeing, and concerns about the reliability of teleconsultation technology, reported by 40.9% of participants. Binary logistic regression revealed that prior usage or exposure to telemedicine was the only significant predictor of positive attitudes, with an odds ratio of 5.688 (95% confidence interval: 1.593-20.313; P = .007).

Conclusion: The inclusion of telemedicine in surgical practice in Jeddah, especially within private healthcare settings, appears promising. The most consistent and significant predictor of positive attitudes toward telemedicine was prior use, as surgeons with previous exposure were more likely to hold favorable views.

Background: Periodontitis is a common chronic inflammatory condition of the supporting tissues of the teeth that destroys the tissues and, if left untreated, results in tooth loss. Accurate and early classification of periodontal bone loss through dental radiographs, such as orthopantomograms (OPGs), is crucial for effective diagnosis and treatment planning.

Objectives: The present study aimed to evaluate and compare 3 deep learning architectures-InceptionV3, InceptionV4, and ResNet-50-for classifying OPGs into distinct grades of dental features characterised by periodontal bone loss.

Design: A comparative experimental design was adopted to analyse the performance of multiple convolutional neural network architectures trained on OPG images representing various grades of periodontal conditions.

Methods: A deep convolutional neural network architecture with varying filter and feature layers was implemented. The training process was conducted using MATLAB on a Dell computer equipped with a GeForce RTX 4060 GPU. Image data augmentation was applied to increase dataset diversity. Several combinations of epochs, learning rates, and optimisation algorithms were tested to enhance performance. Model evaluation metrics included accuracy, precision, recall, and F1-score.

Results: Among the tested architectures, ResNet-50 achieved superior performance, reaching an accuracy of 96.8% by the 16th epoch when trained using an SGD optimiser with momentum and a learning rate of 0.001. It also demonstrated higher precision, recall, and F1 scores compared to InceptionV3 and InceptionV4, confirming its effectiveness in OPG classification.

Conclusion: The findings indicate that ResNet-50 provides better classification accuracy and reliability than InceptionV3 and InceptionV4 in detecting periodontal bone loss from OPG images. Expanding the dataset and exploring advanced data augmentation and hyperparameter tuning could further improve model robustness. This study highlights the potential of deep learning-based OPG classification systems to assist dental professionals in faster and more accurate detection of periodontal diseases.

Osteosarcoma is one of most malignant bone tumors in children, characterized by high recurrence and metastasis. Plumbagin refers to a bioactive compound that is isolated the herb plant of from Plumbagozeylanica zeylanica L., and it has been proven with potential anti-tumor benefits, including osteosarcoma. However, its pharmacological mechanism remains unclear comprehensively. Thus, this study aimed to reveal potential targets and molecular mechanisms in plumbagin for treating osteosarcoma through bioinformatics method and computational validation. In total, respective 379, 2727 and 2166 genes were ascertained as target genes of plumbagin, osteosarcoma and autophagy. A total of 40 shared genes were identified among plumbagin, osteosarcoma and autophagy. Further, additional 10 core genes were identified and used for enrichment analysis. The findings highlighted the regulatory actions of plumbagin on protein-binding, regulation of autophagy for playing anti-osteosarcoma role. Enriched pathway analysis findings disclosed main molecular pathways, including microRNAs in cancer signaling pathway, Notch signaling pathway. Molecular docking data found that the optimal docking affinity and binding energy between plumbagin and scored protein receptors of glycogen synthase kinase 3 beta (GSK3B), histone deacetylase 2 (HDAC2), poly (ADP-ribose) polymerase 1 (PARP1). Our preclinical study investigates the possible therapeutic mechanism of plumbagin against osteosarcoma, indicating that plumbagin exhibited anti-osteosarcoma features via regulation of core target genes associated with autophagy. Current research findings may provide the scientific ideas and evidences for screening bioactive compound against osteosarcoma.

Wearable technology, especially smart glasses, has emerged as a notable breakthrough in healthcare, presenting disruptive possibilities across several domains, including dentistry. Ray-Ban | Meta smart glasses, a cooperation between Meta and EssilorLuxottica, use augmented reality (AR) and artificial intelligence (AI) to improve healthcare operations, patient engagement, and instructional methodologies. This overview maps the possible applications, benefits, difficulties, ethical considerations, and future prospectives of smart glasses in dentistry. This review elucidates how current research indicates that these devices may transform dental practice accuracy, augment education, and boost accessibility, while also tackling issues pertaining to data privacy and ethical use. Overall, smart glasses have the potential to enhance dental education, training, and clinical practice, offering innovative solutions for both educational and practical aspects of dentistry.

Background: Deep learning in brain tumor detection has become an important breakthrough in medical imaging to facilitate a fast and accurate diagnosis. Conventional models such as VGG, SVM, and common CNNs are competitive, yet fail to provide the sensitivity and specificity needed in real-time diagnosis purposes.

Objectives: The objective of the proposed study is to generate a hybrid deep learning architecture of the DenseNet and a self-developed Convolutional Neural Network (CNN) in order to increase the classification accuracy, sensitivity, and specificity of a brain tumor detection medical image.

Design: A Hybrid architecture consisting of DenseNet feature reuse and connectivity with a domain-specific custom CNN is suggested to retrieve high-level semantic and fine-grained features. The design focuses on performance evaluation with respect to the state-of-the-art models and ensemble frameworks.

Methods: The brain tumor data set was preprocessed, and the augmentation methods, such as translation and rotation, were applied. A training and testing subsets of the dataset were formed. The hybrid approach, formed by DenseNet layers and a self-created CNN model, was implemented and tested. The performance of the model was contrasted with that of other benchmark classifiers such as SVM and VGG, DenseNet (single), CNN ensembles, and Hybrid Ensembles.

Results: The hybrid DenseNet-Custom CNN model was more accurate and had better classification results than the standard models. In particular, it surpassed the accuracy of SVM (96%), VGG (94%), DenseNet (92%), and Hybrid Ensemble models (~95.2%), while the FPS remained equivalent to SVM and substantially lower than VGG. It proved better sensitivity and specificity with better feature representation and interpretation, as it made more accurate tumor classification.

Conclusion: Incorporating DenseNet with a custom CNN model can increase the capabilities of brain tumor detection in medical imaging. This hybrid performance can use both general-purpose deep learning and domain-specific feature engineering, and it provides a practical suggestion of the approach that involves a satisfactory solution in the diagnostic sense. It verifies that the combination of more than 2 methods is productive in enhancing the results of medical image classification activities.

Background: Coronary artery disease (CAD) remains one of the leading causes of death globally. Traditional manual scoring methods using non-contrast computed tomography (NCCT) are time-consuming, subjective, and require expertise. To overcome these limitations, this research introduces an AI-driven model to predict and classify more efficiently and accurately. Convolutional Neural Networks (CNNs) are a crucial deep learning tool for detecting cardiovascular diseases (CVDs) from ECG images due to their ability to automatically extract complex patterns and hierarchical features. DenseNet201 is a deep learning model effectively used for cardiovascular disease (CVD) detection from ECG imagery, demonstrating high accuracy in classifying cardiac conditions, particularly for multi-class scenarios. InceptionV3 is a deep learning model widely used for cardiovascular disease (CVD) detection from electrocardiogram (ECG) imagery by leveraging its fine-tuned architecture to classify cardiac conditions.

Objectives: To develop a deep learning-based model for automatic classification and prediction of coronary artery calcium scores. To enhance accuracy using an improved BiGRU model incorporating, to reduce the error and bias in current automatic scoring systems and improve clinical decision-making.

Design: The study designs a novel architecture named HeProbAtt BiGRU Net. The model performs both classification (healthy vs non-healthy) and regression on NCCT image data.

Methods: Data collection, 14 127 NCCT slices-dataset from Tabriz University of Medical Sciences, Preprocessing, Model Development, Performance Evaluation Metrics: Accuracy, precision, recall, F1-score, ROC-AUC, MAE, RMSE.

Results: The proposed model outperformed all compared models with: Classification: Accuracy = 99%, F1-score = 99%, ROC-AUC = .99, Regression: MAE = .065, RMSE = .145. The inclusion of attention and probabilistic weights enhanced learning efficiency and decision precision. Visualization tools (eg, loss curves, confusion matrix, ROC) showed stable and high-performing learning behavior.

Conclusion: The HeProbAtt BiGRU Net provides a highly accurate, automated, and efficient method for coronary artery calcium scoring. Its hybrid framework allows real-time classification and regression, aiding clinicians in early CAD diagnosis. Future work could include validation on larger, multi-center datasets, and incorporation of clinical explain-ability features.

Background: Conventional orthotic insoles demonstrate limited accommodation for individual foot morphology and plantar pressure distribution patterns, resulting in biomechanical inefficiencies and patient discomfort. Computational approaches integrating artificial intelligence with additive manufacturing technologies offer promising solutions for personalized orthotic design. This study investigates the clinical efficacy of AI-driven 3D-printed meshed-silicone orthotics through comprehensive biomechanical assessment.

Methods: A prospective cohort study (n = 21; 8 females, 13 males; age 25.6 ± 3.68 years; BMI 25.48 ± 3.46) evaluated custom orthotics fabricated using machine learning algorithms applied to individual foot and gait data. Pre- and post-intervention assessments included Visual Analog Scale (VAS), Foot Function Index (FFI), Foot Posture Index (FPI), plantar pressure distribution analysis, and 3-dimensional gait analysis over a 4-week period.

Results: FFI scores showed minimal variation (pre-intervention: 13.48 ± 13.14; post-intervention: 14.10 ± 12.96). Significant biomechanical modifications were observed: multi-planar lower extremity alignment correction at hip, knee, and ankle joints. Plantar pressure redistribution demonstrated decreased heel loading with unchanged forefoot pressure distribution, accompanied by significant maximum metatarsal pressure elevation (P < .05).

Conclusions: II. AI-integrated 3D-printed meshed-silicone orthotics demonstrated measurable biomechanical improvements including lower extremity alignment optimization and plantar pressure redistribution. These computational design methodologies combined with advanced manufacturing technologies establish a foundation for personalized orthotic interventions in clinical biomechanics applications.

Background: CT signs of inflammatory and malignant pulmonary nodules are shared and often confused, leading to difficulties in clinical differentiation. Previous relevant studies have neglected to explore the reclassification of morphological signs. This study was designed to evaluate radiomics based on CT images for distinguishing difficult-to-diagnose inflammatory and malignant pulmonary nodules.

Methods: This retrospective study included 333 patients with malignant pulmonary nodules (Mn) and 161 patients with inflammatory pulmonary nodules (In) who were pathologically diagnosed between January 2017 and February 2024. According to whether the CT signs of pulmonary nodules were typical (typical: A or atypical: B), they were further divided into typical malignant nodules (MnA), atypical malignant nodules (MnB), typical inflammatory nodules (InA) and atypical inflammatory nodules (InB). Group 1 (MnA/InA), group 2 (InA/MnB), group 3 (MnA/InB), and group 4 (MnB/InB) were obtained by pairwise comparison. Clinical models, radiomics models and nomogram models were established for each group. The model performance was evaluated by the area under the curve (AUC), accuracy, sensitivity and specificity. The AUCs of the models were compared by using the DeLong test.

Results: In the test set, the AUC values ranged from 0.63 to 0.82. In each group, the nomogram model had the highest diagnostic efficiency and had high accuracy, sensitivity and specificity. For group 3, the nomogram model had the best diagnostic ability (training set: AUC, 0.83; 95% CI [0.75-0.90]; accuracy, 0.72; sensitivity, 0.70; specificity, 0.84, test set: AUC, 0.82; 95% CI [0.70-0.94]; accuracy, 0.65; sensitivity, 0.96).

Conclusions: The nomogram model was useful in diagnosing inflammatory and malignant nodules with typical or atypical signs, especially those with malignant signs, yielding a better classification performance than the radiomics and clinical model.

Background: Operating Room Scheduling (ORS) is vital in healthcare management, impacting patient outcomes, economics, and the shift to value-based care. The academic literature offers various solutions with distinct pros and cons.

Aim: This study aims to (i) outline ORS challenges across surgical specialties; (ii) examine ORS's impact on healthcare goals, focusing on patient outcomes, value-based care, and economics; (iii) assess academic solutions' real-world applicability; and (iv) conduct a bibliometric analysis to track ORS research progression, pivotal works, and future directions.

Methods: We performed a comprehensive bibliometric analysis using Scopus data. Biblioshiny from Bibliometrix aided data mining and analysis, spanning 2000 to 2023, tracking publication trends, themes, co-occurrence, and co-citation networks.

Results: ORS publications steadily rose, notably post-2013, led by developed nations like the UK, Australia, the US, France, and Germany. Key themes included operating rooms, surgery, and humans. Seven primary research routes emerged, covering Surgery Duration, Allocation, Advanced Scheduling Integration, and Patient Flow Optimization. Citation analysis highlighted heuristic algorithms and integer programing as central ORS themes.

Conclusion: This study offers a panoramic ORS overview, advocating an integrated approach aligning patient outcomes, economics, and value-based care. Bibliometric analysis charts ORS research evolution guides future research, and holds significance for practitioners, policymakers, and academics, enhancing ORS paradigms and healthcare delivery.

Tiny Artificial Intelligence (Tiny AI) is transforming resource-constrained embedded systems, particularly in e-health applications, by introducing a shift in Tiny Machine Learning (TinyML) and its integration with the Internet of Things (IoT). Unlike conventional machine learning (ML), which demands substantial processing power, TinyML strategically delegates processing requirements to the cloud infrastructure, allowing lightweight models to run on embedded devices. This study aimed to (i) Develop a TinyML workflow that details the steps for model creation and deployment in resource-constrained environments and (ii) apply the workflow to e-health applications for the real-time detection of epileptic seizures using electroencephalography (EEG) data. The methodology employs a dataset of 4097 EEG recordings per patient, each 23.5 seconds long, from 500 patients, to develop a robust and resilient model. The model was deployed using TinyML on microcontrollers tailored to hardware with limited resources. TensorFlow Lite (TFLite) efficiently runs ML models on small devices, such wearables. Simulation outcomes demonstrated significant performance, particularly in predicting epileptic seizures, with the ExtraTrees Classifier achieving a notable 99.6% Area Under the Curve (AUC) on the validation set. Because of its superior performance, the ExtraTrees Classifier was selected as the preferred model. For the optimized TinyML model, the accuracy remained practically unchanged, whereas inference time was significantly reduced. Additionally, the converted model had a smaller size of 256 KB, approximately ten times smaller, making it suitable for microcontrollers with a capacity of no more than 1 MB. These findings highlight the potential of TinyML to significantly enhance healthcare applications by enabling real-time, energy-efficient decision-making directly on local devices. This is especially valuable in scenarios with limited computing resources or during emergencies, as it reduces latency, ensures privacy, and operates without reliance on cloud infrastructure. Moreover, by reducing the size of training datasets needed, TinyML helps lower overall costs and minimizes the risk of overfitting, making it an even more cost-effective and reliable solution for healthcare innovations.