Biomedical Physics & Engineering Express最新文献

{"title":"Prediction of preoperative tumor-related epilepsy using XGBoost radiomics models with 4 MRI sequences.","authors":"Reuben George, Li Sze Chow, Kheng Seang Lim, Norlisah Ramli, Li Kuo Tan, Mahmud Iwan Solihin","doi":"10.1088/2057-1976/adbdd3","DOIUrl":"10.1088/2057-1976/adbdd3","url":null,"abstract":"<p><p><i>Introduction</i>. Tumor-related epilepsy is a prevalent condition in patients with gliomas. Accurate prediction of epilepsy is crucial for early treatment. This study aimed to evaluate the novel application of the eXtreme Gradient Boost (XGBoost) machine learning (ML) algorithm into a radiomics model predicting preoperative tumor-related epilepsy (PTRE). Its performance was compared with 4 conventional ML algorithms, including the least absolute shrinkage and selection operator (LASSO), elastic net, random forest, and support vector machine.<i>Methods.</i>This study used four magnetic resonance imaging (MRI) images consisting of four sequences (T1-weighted [T1W], T1-weighted contrast [T1WC], T2-weighted [T2W], and T2-weighted fluid-attenuated inversion recovery [T2W FLAIR]) acquired from 74 glioma patients, 30 with PTRE and 44 without PTRE. 394 radiomics features were extracted from the MRI scans using<i>Pyradiomics</i>, alongside 12 clinical features from the medical records. The ML algorithms were mixed and matched to create 20 radiomics models with two stages for: (1) feature selection and (2) prediction of PTRE. Nested cross-validation was used to tune the algorithms and select the stable features.<i>Results.</i>The XGBoost radiomics model demonstrated the second-highest balanced accuracy and F1-score of 0.81 ± 0.01 and 0.80 ± 0.01 respectively. It also achieved the highest recall of 0.81 ± 0.02. It used mostly textural radiomics features from the T1W, T2W and T2W FLAIR sequences to make the predictions.<i>Conclusion.</i>This study demonstrates that XGBoost is a viable alternative to conventional ML algorithms for developing a radiomics model to predict PTRE, as the model produced from XGBoost had among the highest metrics. XGBoost selected features with a higher predictive value than other models. The features selected by XGBoost were more stable, which is a useful property for radiomics analysis. Features selected from multiple MRI sequences were important in the model's decision.</p>","PeriodicalId":8896,"journal":{"name":"Biomedical Physics & Engineering Express","volume":" ","pages":""},"PeriodicalIF":1.3,"publicationDate":"2025-03-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143575687","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":"Design of a compacted-sized eye-tracking system with indexed gaze target for ring gantry linear accelerators: verification study using a motorized eye phantom.","authors":"Queenie T K Shea, April M K Chow, W K Wong, Louis K Y Lee","doi":"10.1088/2057-1976/adbcb0","DOIUrl":"10.1088/2057-1976/adbcb0","url":null,"abstract":"<p><p><i><b>Background</b></i>: Eye-tracking systems and gaze targets provide effective immobilization of the eye during ocular radiation therapy treatment. This study proposes a compact-sized eye-tracking system with a gaze target to provide a simple setup in a ring gantry LINAC. The device should have accurate repositioning, while the gaze target should be easily interchangeable.<i><b>Methods</b></i>: A compact-sized camera was mounted on a 3D-printed stand integrated with a gaze target. The gaze target can be fixed in different positions using an indexed-positioning system. The repositioning accuracy of this system was evaluated using Computed Tomography (CT) scans. The iris center position of the eye phantom was detected using fast radial symmetry transform. The iris center shift in the camera plane was converted to the 3D absolute displacement using the approximation of the spherical rotation of the eye globe. The tolerance of iris motion from gazing toward the gaze target in the camera plane can be estimated with prior knowledge of the iris center location when gazing toward the camera. The detection accuracy was evaluated on a motorized spherical eye phantom.<i><b>Results</b></i>: The mean repositioning errors of the camera and the gaze target were 0.18 mm (Range: 0.08-0.36 mm) and 0.31 mm (Range: 0.03-0.56 mm), respectively. For a typical distance of 200 mm from the camera to the tracked eye, the largest absolute error of iris center detection on a motorized spherical eye phantom was 0.08 mm, while the mean absolute error was less than the resolution of one pixel.<i><b>Conclusions</b></i>: This study proposed a compact-sized eye-tracking system with a gaze target with an easily reproducible setup. By converting the displacement of the motorized eye phantom iris center location in the camera plane to 3D coordinate displacement, accurate tolerance of the gaze direction of the eye can be achieved.</p>","PeriodicalId":8896,"journal":{"name":"Biomedical Physics & Engineering Express","volume":" ","pages":""},"PeriodicalIF":1.3,"publicationDate":"2025-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143565909","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":"FLASH radiotherapy: technical advances, evidence of the FLASH effect and mechanistic insights.","authors":"Mustapha Chaoui, Othmane Bouhali, Yahya Tayalati","doi":"10.1088/2057-1976/adbcb1","DOIUrl":"10.1088/2057-1976/adbcb1","url":null,"abstract":"<p><p>Cancer is one of the leading causes of death worldwide, responsible for nearly 10 million deaths in 2020, with approximately 50% of patients receiving radiation therapy as part of their treatment (Baskar<i>et al</i>2012). Preclinical investigations studies have shown that FLASH radiotherapy (FLASH-RT), delivering radiation in ultra-high dose rates (UHDR), preserves healthy tissue integrity and reduces toxicity, all while maintaining an effective tumor response compared to conventional radiotherapy (CONV-RT), the combined biological benefit was termed as FLASH effect. This article comprehensively surveys pertinent research conducted within FLASH-RT, explores the facilities used in this realm, delves into hypothesized mechanism perspectives, and addresses the challenges to trigger the FLASH effect. In addition, we discuss the potential prospects of FLASH-RT and examine the obstacles that require resolution before its clinical implementation can become a reality.</p>","PeriodicalId":8896,"journal":{"name":"Biomedical Physics & Engineering Express","volume":" ","pages":""},"PeriodicalIF":1.3,"publicationDate":"2025-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143565912","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":"Initial characterization of a novel dual-robot orthovoltage radiotherapy system.","authors":"Olivia Masella, Jacob Atkinson, Sandhya Rottoo, Michael Weil, Magdalena Bazalova-Carter","doi":"10.1088/2057-1976/adbcb2","DOIUrl":"10.1088/2057-1976/adbcb2","url":null,"abstract":"<p><p><b>Purpose:</b>Adequate access to radiotherapy is a critical global concern affecting low-resource settings such as low- and middle-income countries and rural regions. We propose to reduce this disparity by developing a novel low-cost radiotherapy device that treats using non-coplanar techniques and a 225 kVp x-ray tube.<b>Methods:</b>This novel device has been preliminarily characterized spectrally, via spectrometer measurements, dosimetrically, via percent depth dose curves and 2D profiles, and geometrically, via a coplanar star-shot. Dosimetric and geometric evaluations were then combined by performing a proof of workflow of the KOALA system. Monte Carlo simulations were run in TOPAS to validate dosimetric measurements and the proof of workflow measurement.<b>Results:</b>Spectral results showed excellent agreement between measured and modelled spectra. Dose errors of  < 2% were achieved for PDD curves. Full width at half maximum values for the 2D profiles were, on average, 0.95 mm higher in simulation compared to film. A star-shot test demonstrated the high geometrical accuracy of the system with a 0.3 mm diameter wobble circle. Finally, a mean absolute percent error of 5  ±  5% (1<i>σ</i>) was measured for the proof of workflow test.<b>Conclusions:</b>This initial characterization showcased the strengths and weaknesses of the KOALA system, with excellent isocenter precision and depth dose accuracy while lacking dosimetric accuracy in the 2D profiles. Further improvements on the source-to-collimator distance and treatment couch material can be made to improve the accuracy of a Monte Carlo model of the KOALA system.</p>","PeriodicalId":8896,"journal":{"name":"Biomedical Physics & Engineering Express","volume":" ","pages":""},"PeriodicalIF":1.3,"publicationDate":"2025-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143565916","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":"Micromagnetic stimulation (μMS) controls dopamine release: an<i>in vivo</i>study using WINCS<i>Harmoni</i>.","authors":"Renata Saha, Abhinav Goyal, Jason Yuen, Yoonbae Oh, Robert P Bloom, Onri J Benally, Kai Wu, Theoden I Netoff, Walter C Low, Kevin E Bennet, Kendall H Lee, Hojin Shin, Jian-Ping Wang","doi":"10.1088/2057-1976/adbaf9","DOIUrl":"10.1088/2057-1976/adbaf9","url":null,"abstract":"<p><p>Research into the role of neurotransmitters in regulating normal and pathologic brain functions has made significant progress. Yet, clinical trials that aim to improve therapeutic interventions do not take advantage of the<i>in vivo</i>changes in the neurochemistry that occur in real time during disease progression, drug interactions or response to pharmacological, cognitive, behavioral, and neuromodulation therapies. In this work, we used the WINCS<i>Harmoni</i>tool to study the real time<i>in vivo</i>changes in dopamine release in rodent brains for the micromagnetic neuromodulation therapy. Although still in its infancy, micromagnetic stimulation (μMS) using micro-meter sized coils or microcoils (μcoils) has shown incredible promise in spatially selective, galvanic contact free and highly focal neuromodulation. These μcoils are powered by a time-varying current which generates a magnetic field. As per Faraday's Laws of Electromagnetic Induction, this magnetic field induces an electric field in a conducting medium (here, the brain tissues). We used a solenoidal-shaped μcoil to stimulate the medial forebrain bundle (MFB) of the rodent brain<i>in vivo</i>. The evoked<i>in vivo</i>dopamine releases in the striatum were tracked in real time by carbon fiber microelectrodes (CFM) using fast scan cyclic voltammetry (FSCV). Our experiments report that μcoils can successfully activate the MFB in rodent brains, triggering dopamine release<i>in vivo</i>. We further show that the successful release of dopamine upon micromagnetic stimulation is dependent on the orientation of the μcoil. Furthermore, varied intensities of μMS can control the concentration of dopamine releases in the striatum. This work helps us better understand the brain and its conditions arising from a new therapeutic intervention, like μMS, at the level of neurotransmitter release. Despite its early stage, this study potentially paves the path for μMS to enter the clinical world as a precisely controlled and optimized neuromodulation therapy.</p>","PeriodicalId":8896,"journal":{"name":"Biomedical Physics & Engineering Express","volume":" ","pages":""},"PeriodicalIF":1.3,"publicationDate":"2025-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143522445","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":"Exploring the application of deep learning methods for polygenic risk score estimation.","authors":"Steven Squires, Michael N Weedon, Richard A Oram","doi":"10.1088/2057-1976/adbb71","DOIUrl":"10.1088/2057-1976/adbb71","url":null,"abstract":"<p><p><i>Background</i>. Polygenic risk scores (PRS) summarise genetic information into a single number with clinical and research uses. Deep learning (DL) has revolutionised multiple fields, however, the impact of DL on PRSs has been less significant. We explore how DL can improve the generation of PRSs.<i>Methods</i>. We train DL models on known PRSs using UK Biobank data. We explore whether the models can recreate human programmed PRSs, including using a single model to generate multiple PRSs, and DL difficulties in PRS generation. We investigate how DL can compensate for missing data and constraints on performance.<i>Results</i>. We demonstrate almost perfect generation of multiple PRSs with little loss of performance with reduced quantity of training data. For an example set of missing SNPs the DL model produces predictions that enable separation of cases from population samples with an area under the receiver operating characteristic curve of 0.847 (95% CI: 0.828-0.864) compared to 0.798 (95% CI: 0.779-0.818) for the PRS.<i>Conclusions</i>. DL can accurately generate PRSs, including with one model for multiple PRSs. The models are transferable and have high longevity. With certain missing SNPs the DL models can improve on PRS generation; further improvements would likely require additional input data.</p>","PeriodicalId":8896,"journal":{"name":"Biomedical Physics & Engineering Express","volume":" ","pages":""},"PeriodicalIF":1.3,"publicationDate":"2025-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143527857","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":"Morphometric variation in central airways of ten different human lung.","authors":"M Monfared, M Mohammadzadeheydgahi, Z Farshidfar, S Mehrabi, S Sadrizadeh, O Abouali","doi":"10.1088/2057-1976/adbbf4","DOIUrl":"10.1088/2057-1976/adbbf4","url":null,"abstract":"<p><p>The prevailing scarcity of accurate lung models poses challenges to predicting airborne particle deposition across genders. The present work demonstrates the details of the geometrical specifications of central airways for ten healthy humans (male and female). The data were extracted from HRCT scan images with a minimum resolution of 1 mm. The images cover the trachea to all branches of the G6-G8 generations. The presented data include airway segment diameters, lengths, branching angles, and angles of inclination to gravity, in addition to their average and standard deviation. Our first goal in this study is to generate an average lung model exclusively for humans in laboratory and 1D numerical inhalation investigations. Thus, our primary emphasis in this work is to find the average suitable inclination angle in all generations of central airways for men and women by comparing the available data from previous studies. In the second part of the paper, we have also investigated the particle deposition efficiency in these ten models using the Mimetikos PreludiumTM software package. We compared the regional deposition between males and females and the available respiratory system models.</p>","PeriodicalId":8896,"journal":{"name":"Biomedical Physics & Engineering Express","volume":" ","pages":""},"PeriodicalIF":1.3,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143540034","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":"Reconstruction of ECG from ballistocardiogram using generative adversarial networks with attention.","authors":"Hong Chen, Jing Zhan, Ruilin Feng, Kewei Chen, Tao Zhao, Xuelei Fu, Zhengying Li","doi":"10.1088/2057-1976/adbbf5","DOIUrl":"10.1088/2057-1976/adbbf5","url":null,"abstract":"<p><p>Electrocardiogram (ECG) is widely used to provide early warning signals for cardiovascular diseases. However, traditional twelve-lead ECG monitoring methods and smartwatch-based home solutions are unable to achieve daily long-term monitoring. Therefore, in this work, we propose a system to reconstruct ECG signals from non-contact Ballistocardiogram (BCG) signals. First, we synchronously collect BCG and ECG signals using fiber optic sensors and an ECG machine, and preprocess the signals to obtain a training set. We train the Att-SNGAN model using this training set to reconstruct ECG signals from BCG inputs. Experimental results show that the reconstructed ECG signals have a mean absolute error (MAE) of only 0.0651, a Root Mean Square Error (RMSE) of 0.0735 and a Fréchet Distance (FD) of 0.0342, showing high consistency with the original ECG. This work highlights the significant potential of the system for continuous cardiac cycle monitoring and HRV analysis, providing new solutions for long-term ECG monitoring at home.</p>","PeriodicalId":8896,"journal":{"name":"Biomedical Physics & Engineering Express","volume":" ","pages":""},"PeriodicalIF":1.3,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143540036","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":"Hierarchical agent transformer network for COVID-19 infection segmentation.","authors":"Yi Tian, Qi Mao, Wenfeng Wang, Yan Zhang","doi":"10.1088/2057-1976/adbafa","DOIUrl":"10.1088/2057-1976/adbafa","url":null,"abstract":"<p><p>Accurate and timely segmentation of COVID-19 infection regions is critical for effective diagnosis and treatment. While convolutional neural networks (CNNs) exhibit strong performance in medical image segmentation, they face challenges in handling complex lesion morphologies with irregular boundaries. Transformer-based approaches, though demonstrating superior capability in capturing global context, suffer from high computational costs and suboptimal multi-scale feature integration. To address these limitations, we proposed Hierarchical Agent Transformer Network (HATNet), a hierarchical encoder-bridge-decoder architecture that optimally balances segmentation accuracy with computational efficiency. The encoder employs novel agent Transformer blocks specifically designed to capture subtle features of small COVID-19 lesions through agent tokens with linear computational complexity. A diversity restoration module (DRM) is innovatively embedded within each agent Transformer block to counteract feature degradation. The hierarchical structure simultaneously extracts high-resolution shallow features and low-resolution fine features, ensuring comprehensive feature representation. The bridge stage incorporates an improved pyramid pooling module (IPPM) that establishes hierarchical global priors, significantly improving contextual understanding for the decoder. The decoder integrates a full-scale bidirectional feature pyramid network (FsBiFPN) with a dedicated border-refinement module (BRM), collectively enhancing edge precision. The HATNet were evaluated on the COVID-19-CT-Seg and CC-CCII datasets. Experimental results yielded Dice scores of 84.14% and 81.22% respectively, demonstrating superior segmentation performance compared to state-of-the-art models. Furthermore, it achieved notable advantages in model parameters and computational complexity, highlighting its clinical deployment potential.</p>","PeriodicalId":8896,"journal":{"name":"Biomedical Physics & Engineering Express","volume":" ","pages":""},"PeriodicalIF":1.3,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143522303","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":"Evaluation of image quality and fetal dose through current modulation CT-scan using pregnancy phantoms.","authors":"Endarko Endarko, Fathul Jannah, Aditya Prayugo Hariyanto, Isfina Uniatunada, Choirul Anam, Nurhanifa Tri Budiarti","doi":"10.1088/2057-1976/adb9ed","DOIUrl":"10.1088/2057-1976/adb9ed","url":null,"abstract":"<p><p>Computed Tomography (CT) is often required in special circumstances during pregnancy to determine internal medicine, specifically when information from other imaging techniques, such as ultrasound and magnetic resonance imaging (MRI), is still inadequate. In CT-Chest, abdomen, and pelvis (CAP) examination, direct irradiation of the fetal is necessary, indicating that the potential for fetal exposure must be considered. Therefore, this study evaluated the effects of current modulation on image quality and fetal absorbed dose in pregnancy CT scan. Calculation using IndoseCT and film dosimeters were used with a 3D-printed anthropomorphic pregnant phantom thorax-abdomen-pelvic during the first and third trimesters of pregnancy. Image quality analysis and image noise were then measured by IndoseCT, while fetal dose analysis was performed using IndoseCT software, as well as through direct measurements with an XR-QA2 film dosimeter. Statistical tests were performed to compare the data obtained using both methods. The results showed that the use of current modulation increased the image noise. The fetal dose can be significantly reduced by adjusting the tube current and patient diameter without affecting image quality in the first and third trimesters. It was also revealed that the smallest diameter received the largest dose in both trimesters for nonmodulation. Although there were differences in the dose values obtained from the IndoseCT and measurements, the data patterns were not significantly different. Furthermore, based on these results, the dosage value was below the tolerance threshold for deterministic effects (i.e., <50 mGy).</p>","PeriodicalId":8896,"journal":{"name":"Biomedical Physics & Engineering Express","volume":" ","pages":""},"PeriodicalIF":1.3,"publicationDate":"2025-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143498526","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}