Biomedical Physics & Engineering Express最新文献

{"title":"Monte Carlo investigation of dosimetry in shielded region in total body irradiation treatment.","authors":"Peixiong Li, Xiaomin Liang, Qiuwen Wu","doi":"10.1088/2057-1976/adf3b6","DOIUrl":"10.1088/2057-1976/adf3b6","url":null,"abstract":"<p><p><i>Purpose</i>. Total body irradiation (TBI) is commonly used to treat hematological diseases requiring bone marrow transplantation. Partial transmission blocks (PTB) are utilized to shield critical organs such as the lungs and kidneys. Previous phantom measurement and convolution algorithms confirmed that the percent depth dose (PDD) under PTB deviates significantly from those regions without the PTB. In this study, we investigated the dosimetry under the PTB using the Monte Carlo tool and validated with measurement.<i>Methods</i>. The photon phase space (PSP) for Truebeam linac from MyVarian was used as input in the EGSnrc package. The PSP was analyzed and separated into primary point source photons (originating from the target) and extra-focal source photons (extra-focal source originating from flattening filter etc) components. It was hypothesized that they behave differently in the presence of PTB which is responsible for the uncommon dosimetry. A virtual filter module was developed to simulate the PTB of any transmission factors in EGSnrc. Further, a concept of Extra-focal Source Photon Enhancement Ratio (ESPER) was proposed to quantify how the extra-focal source photons' dose contribution changes with SSD, block size, block-to-patient distance, and transmission factor.<i>Results</i>. Extra-focal source photons accounts for 20% in 6X beam, but only 13% in 6XFFF. MC result of the virtual PTB filter agrees well with the measurement for PDD (<1.5%). For a clinical PTB of size 6 × 12 cm²at phantom surface, the ESPER at 5 cm depth increases from 1.34 to 2.38 when SSD 100 → 400 cm; decreases from 1.96 to 1.07 when block-surface-distance 149.7 → 10 cm; and decreases from 4.00 to 1.65 when PTB transmission factor 0 → 30%.<i>Conclusions</i>. The dosimetry under PTB for TBI can be explained by the different behavior of the primary point source and extra focal source photon components in phantom. This study provides valuable inputs on how to interpret the<i>in vivo</i>dose measurement under PTB.</p>","PeriodicalId":8896,"journal":{"name":"Biomedical Physics & Engineering Express","volume":" ","pages":""},"PeriodicalIF":1.6,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144706151","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A lightweight hybrid DL model for multi-class chest x-ray classification for pulmonary diseases.","authors":"J Glory Precious, Rajkumar S, Sarayu Pratika B, Vaisnav R R, Syed Sheik Mohamed M, V Sapthagirivasan","doi":"10.1088/2057-1976/adf3b8","DOIUrl":"10.1088/2057-1976/adf3b8","url":null,"abstract":"<p><p>Pulmonary diseases have become one of the main reasons for people's health decline, impacting millions of people worldwide. Rapid advancement of deep learning has significantly impacted medical image analysis by improving diagnostic accuracy and efficiency. Timely and precise diagnosis of these diseases proves to be invaluable for effective treatment procedures. Chest x-rays (CXR) perform a pivotal role in diagnosing various respiratory diseases by offering valuable insights into the chest and lung regions. This study puts forth a hybrid approach for classifying CXR images into four classes namely COVID-19, tuberculosis, pneumonia, and normal (healthy) cases. The presented method integrates a machine learning method, Support Vector Machine (SVM), with a pre-trained deep learning model for improved classification accuracy and reduced training time. Data from a number of public sources was used in this study, which represents a wide range of demographics. Class weights were implemented during training to balance the contribution of each class in order to address the class imbalance. Several pre-trained architectures, namely DenseNet, MobileNet, EfficientNetB0, and EfficientNetB3, have been investigated, and their performance was evaluated. Since MobileNet achieved the best classification accuracy of 94%, it was opted for the hybrid model, which combines MobileNet with SVM classifier, increasing the accuracy to 97%. The results suggest that this approach is reliable and holds great promise for clinical applications.</p>","PeriodicalId":8896,"journal":{"name":"Biomedical Physics & Engineering Express","volume":" ","pages":""},"PeriodicalIF":1.6,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144706130","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Subject-specific feature extraction approach for a three-class motor imagery-based brain-computer interface enabling navigation in a virtual environment: open-access framework.","authors":"Fardin Afdideh, Mohammad Bagher Shamsollahi","doi":"10.1088/2057-1976/aded19","DOIUrl":"10.1088/2057-1976/aded19","url":null,"abstract":"<p><p>Brain-Computer Interface (BCI) is a system that aids individuals with disabilities to establish a novel communication channel between the brain and computer. Among various electrophysiological sources that can drive a BCI system, Motor Imagery (MI) facilitates more natural communication for users with motor disabilities, whereas electroencephalogram (EEG) is considered the most practical brain imaging modality. However, subject training is a critical aspect of such a type of BCI. One possible solution to address this challenge is to leverage the Virtual Reality (VR) technology. This study proposes a VR in MI- and EEG-based BCI (MI-EEG-BCI-VR) framework wherein users navigate a Virtual Environment (VE) following cue-based training, and employing a subject-specific feature extraction approach. The assigned task involves performing the left hand, right hand, and feet movement imagination to navigate from the start station to the end station as quickly as possible. The generated brain signals are collected using three bipolar EEG channels only. The proposed open-access MATLAB-based MI-EEG-BCI-VR framework was validated with eight healthy participants. One participant demonstrated satisfactory performance in navigating the VE. Notably, it achieved the highest performance of 82.28 ± 5.11% for MI and 97.72 ± 4.55% for Motor Execution (ME) after just a single training session.</p>","PeriodicalId":8896,"journal":{"name":"Biomedical Physics & Engineering Express","volume":" ","pages":""},"PeriodicalIF":1.6,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144590358","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhancing surface electromyographic signal recognition accuracy for trans-radial amputees using broad learning systems.","authors":"Lei Zhang, Xuemei Zhang","doi":"10.1088/2057-1976/adee28","DOIUrl":"10.1088/2057-1976/adee28","url":null,"abstract":"<p><p>Gesture recognition based on surface electromyography (sEMG) plays a crucial role in human-computer interaction. By analyzing sEMG signals generated from residual forearm muscle activity in trans-radial amputees, it is possible to predict their hand movement intentions, enabling the control of myoelectric prostheses. Previous studies on gesture classification for trans-radial amputees have shown that as the number of hand movement types increases, the accuracy of gesture classification significantly decreases. To this end, this paper proposed a novel approach that integrates image feature flattening (IFF) with broad learning system (BLS). The IFF method mapped data features into a three-dimensional image, which was then converted to grayscale and flattened into a one-dimensional vector. Finally, it was used as input to the BLS network for precise gesture classification. The proposed method was validated on 49 hand movement data from radial amputees in the Ninapro DB3 dataset. The results showed that the method not only significantly reduced classification time but also achieved a gesture recognition accuracy of up to 98.1%, demonstrating its strong potential for application in gesture recognition for trans-radial amputees.</p>","PeriodicalId":8896,"journal":{"name":"Biomedical Physics & Engineering Express","volume":" ","pages":""},"PeriodicalIF":1.6,"publicationDate":"2025-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144607189","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-modal classification of retinal disease based on convolutional neural network.","authors":"Hongyi Pan, Jingpeng Miao, Jie Yu, Jingmin Li, Xiaobing Wang, Jihong Feng","doi":"10.1088/2057-1976/adeb92","DOIUrl":"10.1088/2057-1976/adeb92","url":null,"abstract":"<p><p>Retinal diseases such as age-related macular degeneration and diabetic retinopathy will lead to irreversible blindness without timely diagnosis and treatment. Optical coherence tomography (OCT) and optical coherence tomography angiography (OCTA) images provide complementary views of the retina, and the integration of the two imaging modalities can improve the accuracy of retinal disease classification. We propose a multi-modal classification model consisting of two branches to automatically diagnose retinal diseases, in which OCT and OCTA images are efficiently integrated to improve both the accuracy and efficiency of disease diagnosis. A bright line cropping is used to remove the useless black edge region while preserving the lesion features and reducing the calculation load. To solve the insufficient data issue, data enhancement and loose matching methods are adopted to increase the data amount. A two-step training method is used to train our proposed model, alleviating the limited training images. Our model is tested on an external test set instead of a training set, making the classification results more rigorous. The intermediate fusion and two-step training methods are adopted in our multiple classification model, achieving 0.9667, 0.9418, 0.8569, 0.9422, and 0.8921 in average accuracy, precision, recall, specificity, and F1-Score, respectively. Our multi-modal model outperforms the single-modal model, the early, and late fusion multi-modal model in accuracy. Our model offers doctors less human error, lower cost, more uniform, and effective mass screening, thus providing a solution to improve deep learning performance in terms of a relatively fewer number of training data and even more imbalanced classes.</p>","PeriodicalId":8896,"journal":{"name":"Biomedical Physics & Engineering Express","volume":" ","pages":""},"PeriodicalIF":1.6,"publicationDate":"2025-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144558927","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The proton therapy research beamline at the Christie NHS foundation trust.","authors":"Nicholas T Henthorn, John-William Warmenhoven, Samuel P Ingram, Samuel P Manger, Michael J Merchant, Hywel Owen, Ranald I Mackay, Karen J Kirkby, Michael J Taylor","doi":"10.1088/2057-1976/addbe8","DOIUrl":"10.1088/2057-1976/addbe8","url":null,"abstract":"<p><p>Proton therapy is a relatively new modality for cancer treatment and has several open research questions, particularly in the biological realm. Due to large infrastructure costs the modality is reserved for specialist treatment, limiting the patient outcome dataset. This requires supplementation with fundamental research through<i>in vitro</i>and<i>in vivo</i>systems. Similarly, the safety and potential benefits of new treatments, such as FLASH, should be demonstrated in lab environments prior to clinical translation. Greater access to clinically relevant research platforms is required. This work presents the capabilities of the Manchester proton therapy research facility for experimentalists' assessment to meet their research goals. Details of the research beamline geometry are presented, along with workflows for<i>in vitro</i>sample irradiation within an automated sample handling environmental chamber. Absolute dose and dose depth of the proton research beamline was measured. The dose calibration across a range of energies and dose rates is presented and fits are mathematically described. Methods to convert measured, or planned, dose to sample dose are presented including for biological studies investigating end of proton range effects. Elements of the beam optics, impacting on spot size and therefore field homogeneity, were measured for sample irradiation and beam model development. A Monte Carlo beam model was established to predict physically difficult measurements and is compared to measurements throughout. Achievable dose rates for FLASH are presented alongside absolute dosimetric accuracy. There was a focus on radiobiological research in establishing the beamline. Special care was taken to develop high-throughput repeatable<i>in vitro</i>irradiation workflows, with an adjacent radiobiological lab for immediate processing. This will lead to a reduction in experimental uncertainties seen in the literature with demonstrated accurate dosimetry, tight environmental control, and a high degree of versatility. The infrastructure presented in this work is a unique facility in the UK.</p>","PeriodicalId":8896,"journal":{"name":"Biomedical Physics & Engineering Express","volume":" ","pages":""},"PeriodicalIF":1.6,"publicationDate":"2025-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12296266/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144126395","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Recent advances in applying machine learning to proton radiotherapy.","authors":"Vanessa L Wildman, Jacob F Wynne, Shadab Momin, Aparna H Kesarwala, Xiaofeng Yang","doi":"10.1088/2057-1976/adeb90","DOIUrl":"10.1088/2057-1976/adeb90","url":null,"abstract":"<p><p><i>Background</i>.<i>Objectives</i>: In radiation oncology, precision and timeliness of both planning and treatment are paramount values of patient care. Machine learning has increasingly been applied to various aspects of photon radiotherapy to reduce manual error and improve the efficiency of clinical decision making; however, applications to proton therapy remain an emerging field in comparison. This systematic review aims to comprehensively cover all current and potential applications of machine learning to the proton therapy clinical workflow, an area that has not been extensively explored in literature.<i>Methods</i>: PubMed and Embase were utilized to identify studies pertinent to machine learning in proton therapy between 2019 to 2024. An initial search on PubMed was made with the search strategy ''proton therapy', 'machine learning', 'deep learning''. A subsequent search on Embase was made with '('proton therapy') AND ('machine learning' OR 'deep learning')'. In total, 38 relevant studies have been summarized and incorporated.<i>Results</i>: It is observed that U-Net architectures are prevalent in the patient pre-screening process, while convolutional neural networks play an important role in dose and range prediction. Both image quality improvement and transformation between modalities to decrease extraneous radiation are popular targets of various models. To adaptively improve treatments, advanced architectures such as general deep inception or deep cascaded convolution neural networks improve online dose verification and range monitoring.<i>Conclusions</i>: With the rising clinical usage of proton therapy, machine learning models have been increasingly proposed to facilitate both treatment and discovery. Significantly improving patient screening, planning, image quality, and dose and range calculation, machine learning is advancing the precision and personalization of proton therapy.</p>","PeriodicalId":8896,"journal":{"name":"Biomedical Physics & Engineering Express","volume":" ","pages":""},"PeriodicalIF":1.3,"publicationDate":"2025-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12284894/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144558928","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Development and validation of an improved volumetric breast density estimation model using the ResNet technique.","authors":"Yoshiyuki Asai, Mika Yamamuro, Takahiro Yamada, Yuichi Kimura, Kazunari Ishii, Yusuke Nakamura, Yujiro Otsuka, Yohan Kondo","doi":"10.1088/2057-1976/adecac","DOIUrl":"10.1088/2057-1976/adecac","url":null,"abstract":"<p><p><i>Objective</i>. Temporal changes in volumetric breast density (VBD) may serve as prognostic biomarkers for predicting the risk of future breast cancer development. However, accurately measuring VBD from archived x-ray mammograms remains challenging. In a previous study, we proposed a method to estimate volumetric breast density using imaging parameters (tube voltage, tube current, and exposure time) and patient age. This approach, based on a multiple regression model, achieved a determination coefficient (R²) of 0.868.<i>Approach</i>. In this study, we developed and applied machine learning models-Random Forest, XG-Boost-and the deep learning model Residual Network (ResNet) to the same dataset. Model performance was assessed using several metrics: determination coefficient, correlation coefficient, root mean square error, mean absolute error, root mean square percentage error, and mean absolute percentage error. A five-fold cross-validation was conducted to ensure robust validation.<i>Main results</i>. The best-performing fold resulted in R²values of 0.895, 0.907, and 0.918 for Random Forest, XG-Boost, and ResNet, respectively, all surpassing the previous study's results. ResNet consistently achieved the lowest error values across all metrics.<i>Significance</i>. These findings suggest that ResNet successfully achieved the task of accurately determining VBD from past mammography-a task that has not been realised to date. We are confident that this achievement contributes to advancing research aimed at predicting future risks of breast cancer development by enabling high-accuracy time-series analyses of retrospective VBD.</p>","PeriodicalId":8896,"journal":{"name":"Biomedical Physics & Engineering Express","volume":" ","pages":""},"PeriodicalIF":1.3,"publicationDate":"2025-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144582961","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Polydopamine-modified collagen membrane loading with platelet-rich plasma for enhancing diabetic wound healing.","authors":"Hao-Jie Gao, Xiao-Wan Fang, Hao Chen, Zhen-Zhen Yan, Fei Xu, Chao Ji, Zi-Xuan Zhou, Yu-Xiang Wang, Jing-Nan Xun, Yi-Xin Wu, Fu-Ting Shu, Yong-Jun Zheng, Shi-Chu Xiao","doi":"10.1088/2057-1976/adebf6","DOIUrl":"10.1088/2057-1976/adebf6","url":null,"abstract":"<p><p>Platelet-rich plasma (PRP), a reservoir of growth factors, is instrumental in the repair and regeneration of damaged tissues, orchestrating wound healing at all stages. However, PRP's rapid degradation and instability at the wound site, prone to displacement and degradation, limit its efficacy. Collagen, the most abundant protein in the human body, boasts exceptional biocompatibility, biological activity, and minimal immunogenicity. Polydopamine (PDA)-coated materials have been employed for sustained drug release, leveraging the catechol, amine, and imine functional groups on their surface for covalent bonding with other molecules. This study presents the fabrication of a PDA-modified collagen membrane (PDA-CM) loaded with PRP (PDA-CM@PRP) to achieve a sustained release of PRP. Our results showed that PDA-CM@PRP significantly improved proliferation, migration, delayed cellular senescence and reduced oxidative stress in human dermal fibroblasts (HDFs)<i>in vitro</i>.<i>In vivo</i>experiments demonstrated accelerated diabetic wound healing with enhanced granulation tissue formation, cell proliferation, and neovascularization. Transcriptome sequencing analysis revealed that PDA-CM@PRP activated HDFs proliferation through upregulation of the cell cycle and DNA replication pathways. This study presents a novel strategy for sustained PRP release, offering a promising therapeutic approach for diabetic wounds and other chronic wound types.</p>","PeriodicalId":8896,"journal":{"name":"Biomedical Physics & Engineering Express","volume":" ","pages":""},"PeriodicalIF":1.3,"publicationDate":"2025-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144564325","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"ViT-GCN: a novel hybrid model for accurate pneumonia diagnosis from x-ray images.","authors":"Nuo Xu, Jinran Wu, Fengjing Cai, Xi'an Li, Hong-Bo Xie","doi":"10.1088/2057-1976/adebf4","DOIUrl":"10.1088/2057-1976/adebf4","url":null,"abstract":"<p><p>This study aims to enhance the accuracy of pneumonia diagnosis from x-ray images by developing a model that integrates Vision Transformer (ViT) and Graph Convolutional Networks (GCN) for improved feature extraction and diagnostic performance. The ViT-GCN model was designed to leverage the strengths of both ViT, which captures global image information by dividing the image into fixed-size patches and processing them in sequence, and GCN, which captures node features and relationships through message passing and aggregation in graph data. A composite loss function combining multivariate cross-entropy, focal loss, and GHM loss was introduced to address dataset imbalance and improve training efficiency on small datasets. The ViT-GCN model demonstrated superior performance, achieving an accuracy of 91.43% on the COVID-19 chest x-ray database, surpassing existing models in diagnostic accuracy for pneumonia. The study highlights the effectiveness of combining ViT and GCN architectures in medical image diagnosis, particularly in addressing challenges related to small datasets. This approach can lead to more accurate and efficient pneumonia diagnoses, especially in resource-constrained settings where small datasets are common.</p>","PeriodicalId":8896,"journal":{"name":"Biomedical Physics & Engineering Express","volume":" ","pages":""},"PeriodicalIF":1.3,"publicationDate":"2025-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144564326","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}