Physical and Engineering Sciences in Medicine最新文献

{"title":"An approach for cancer outcomes modelling using a comprehensive synthetic dataset.","authors":"Lorna Tu, Hervé H F Choi, Haley Clark, Samantha A M Lloyd","doi":"10.1007/s13246-025-01594-2","DOIUrl":"10.1007/s13246-025-01594-2","url":null,"abstract":"<p><p>Limited patient data availability presents a challenge for efficient machine learning (ML) model development. Recent studies have proposed methods to generate synthetic medical images but lack the corresponding prognostic information required for predicting outcomes. We present a cancer outcomes modelling approach that involves generating a comprehensive synthetic dataset which can accurately mimic a real dataset. A real public dataset containing computed tomography-based radiomic features and clinical information for 132 non-small cell lung cancer patients was used. A synthetic dataset of virtual patients was synthesized using a conditional tabular generative adversarial network. Models to predict two-year overall survival were trained on real or synthetic data using combinations of four feature selection methods (mutual information, ANOVA F-test, recursive feature elimination, random forest (RF) importance weights) and six ML algorithms (RF, k-nearest neighbours, logistic regression, support vector machine, XGBoost, Gaussian Naïve Bayes). Models were tested on withheld real data and externally validated. Real and synthetic datasets were similar, with an average one minus Kolmogorov-Smirnov test statistic of 0.871 for continuous features. Chi-square test confirmed agreement for discrete features (p < 0.001). XGBoost using RF importance-based features performed the most consistently for both datasets, with percent differences in balanced accuracy and area under the precision-recall curve of < 1.3%. Preliminary findings demonstrate the potential application of synthetic radiomic and clinical data augmentation for cancer outcomes modelling, although further validation with larger diverse datasets is crucial. While our approach was described in a lung context, it may be applied to other sites or endpoints.</p>","PeriodicalId":48490,"journal":{"name":"Physical and Engineering Sciences in Medicine","volume":" ","pages":"1473-1483"},"PeriodicalIF":2.0,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144700047","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The future of scientific publishing: challenges and a vision for change.","authors":"Clive Baldock","doi":"10.1007/s13246-025-01642-x","DOIUrl":"10.1007/s13246-025-01642-x","url":null,"abstract":"","PeriodicalId":48490,"journal":{"name":"Physical and Engineering Sciences in Medicine","volume":" ","pages":"961-962"},"PeriodicalIF":2.0,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145066141","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluation of volumetric modulated arc therapy dose calculation plans based on kV-cone beam CT images for the breast cancer treatment.","authors":"Amine El Outmani, M Zerfaoui, I Hattal, Y Oulhouq, K Bahhous, A Rrhioua, D Bakari, M Hamal, A Moussa","doi":"10.1007/s13246-025-01592-4","DOIUrl":"10.1007/s13246-025-01592-4","url":null,"abstract":"<p><p>The majority of treatment centers currently use Cone Beam Computed Tomography (CBCT) as an effective method of patient repositioning, opening up the prospect of using it in treatment plans calculation. The purpose of this task is to investigate the feasibility of utilizing kV-CBCT images as a potential alternative to CT scans for generating treatment plans during radiation therapy. The images are taken by the imaging system installed on a linear accelerator and implemented in a Treatment Planning System (TPS). This imaging system opens the possibility to proceed to dose calculation using a calibration curve that establishes a link between voxel attenuation in Hounsfield Units (HU) and electron densities relative to water (RED). Furthermore, a comparison of Volumetric Modulated Arc Therapy (VMAT) treatment plans on this imaging modality was done to the Computed Tomography (CT) imaging system using the CIRS phantom. The comparison is also performed the images of 10 patients with right breast cancer. VMAT plans from the two modalities were compared in terms of target coverage, normal tissue sparing, dose distribution parameters, and monitor units (MUs). Then, the gamma index test is employed by using the PTW Verisoft to compare the TPS calculated dose distribution for the two modalities. It is also used to compare the measured dose distribution performed by the Portal Dosimetry. Regarding the volume-dose parameters of the PTV and the Organs at risk (OARs), no differences were found between the VMAT plans of the two imaging modalities, CT and CBCT. Additionally, the gamma analysis results of the patients VMAT plans for 1%-1 mm, 2%-2 mm, 3%-3 mm, 4%-4 mm and 5%-5 mm showed that more than 84%, 90%, 92%, 94%, and 96%, respectively, of the points agreed upon between Eclipse calculated and measured dose distributions for the CT and CBCT VMAT plans. The good dosimetric agreement (gamma index reaches more than 92% for 3%-3 mm) between breast VMAT plans based on CT and CBCT images renders the latter an appealing verification tool and a substitute for CT images, if needed, for dose calculation. CBCT images are an effective option for direct dose calculation or to do adaptive radiotherapy in the case of a breast cancer.</p>","PeriodicalId":48490,"journal":{"name":"Physical and Engineering Sciences in Medicine","volume":" ","pages":"1389-1398"},"PeriodicalIF":2.0,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144676127","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An enhanced deep learning framework for muscle artifact removal from ECG signal integrating resnet, GCAB, and BI-LSTM.","authors":"Pavan G Malghan, Malaya Kumar Hota","doi":"10.1007/s13246-025-01584-4","DOIUrl":"10.1007/s13246-025-01584-4","url":null,"abstract":"<p><p>Electrocardiogram (ECG) signals are significantly distorted during recording by muscle artifact (MA), causing signal frequency overlap and making it difficult to interpret ECG data correctly. Deep learning (DL) methods for signal processing have shown promising results. However, there is a significant necessity in building proper DL models with appropriate datasets. We propose an enhanced hybrid deep learning framework called HRGB-Net based on residual neural network (ResNet), global channel attention block (GCAB), and bidirectional-long-short-term memory (Bi-LSTM) blocks for filtering the MA noise from ECG by using three distinctive MIT-BIH real-time datasets from the PhysioNet repository by creating suitable datasets for training. We use both raw ECG data and short-time Fourier-transformed (STFT) ECG data for comparative analysis with three neural network models: a convolutional neural Network (CNN), a fully connected neural network (FCNN), and a regression-based LSTM (Reg-LSTM-DNN) model to assess the proposed model. The signal-to-noise ratio (SNR) of noisy ECG signals is varied from - 7dB to 2dB to analyze the mean square error (MSE) and correlation coefficient (CC) performances after the denoising process. Our proposed method utilizes the regression ability to remove MA noise and generate a clean ECG signal with improved values of these signal parameters. The STFT-trained and tested ECG data shows better results than the raw ECG data for efficiently eliminating the MA with a 98.82% correlation coefficient and optimal MSE value of 0.053068. The results prove our proposed HRGB-Net model's remarkable ability to outperform the neural network models and other standard techniques.</p>","PeriodicalId":48490,"journal":{"name":"Physical and Engineering Sciences in Medicine","volume":" ","pages":"1281-1298"},"PeriodicalIF":2.0,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144477405","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A novel multimodal computer-aided diagnostic model for pulmonary embolism based on hybrid transformer-CNN and tabular transformer.","authors":"Wei Zhang, Yu Gu, Hao Ma, Lidong Yang, Baohua Zhang, Jing Wang, Meng Chen, Xiaoqi Lu, Jianjun Li, Xin Liu, Dahua Yu, Ying Zhao, Siyuan Tang, Qun He","doi":"10.1007/s13246-025-01568-4","DOIUrl":"10.1007/s13246-025-01568-4","url":null,"abstract":"<p><p>Pulmonary embolism (PE) is a life-threatening clinical problem where early diagnosis and prompt treatment are essential to reducing morbidity and mortality. While the combination of CT images and electronic health records (EHR) can help improve computer-aided diagnosis, there are many challenges that need to be addressed. The primary objective of this study is to leverage both 3D CT images and EHR data to improve PE diagnosis. First, for 3D CT images, we propose a network combining Swin Transformers with 3D CNNs, enhanced by a Multi-Scale Feature Fusion (MSFF) module to address fusion challenges between different encoders. Secondly, we introduce a Polarized Self-Attention (PSA) module to enhance the attention mechanism within the 3D CNN. And then, for EHR data, we design the Tabular Transformer for effective feature extraction. Finally, we design and evaluate three multimodal attention fusion modules to integrate CT and EHR features, selecting the most effective one for final fusion. Experimental results on the RadFusion dataset demonstrate that our model significantly outperforms existing state-of-the-art methods, achieving an AUROC of 0.971, an F1 score of 0.926, and an accuracy of 0.920. These results underscore the effectiveness and innovation of our multimodal approach in advancing PE diagnosis.</p>","PeriodicalId":48490,"journal":{"name":"Physical and Engineering Sciences in Medicine","volume":" ","pages":"1107-1126"},"PeriodicalIF":2.0,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144136388","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comparative characterisation of different types of Gafchromic films for radiotherapy use.","authors":"Tarafder Shameem, Nick Bennie, Martin Butson, David Thwaites","doi":"10.1007/s13246-025-01596-0","DOIUrl":"10.1007/s13246-025-01596-0","url":null,"abstract":"<p><p>Different types of Gafchromic films, for radiotherapy use, are recommended for different dose ranges. Ashland Specialty Ingredients has aimed to continuously develop its products to improve their practical application. Thus, EBT3 was replaced by EBT4, intended to provide better signal to noise ratio (SNR); while MD-V3 was introduced for use at higher dose ranges, in addition to EBT-XD. At present there are limited studies on MD-V3. This study aimed to investigate some relevant characteristics of EBT4 and MD-V3, compared with those of EBT3 and EBT-XD. The parameters investigated were dose response, optical density change with post-irradiation time, orientation effect, signal to noise ratio, polarisation, and lateral response artefact (LRA). EBT4 is similar to EBT3 however it provides better SNR and larger response change with post-irradiation time. EBT-XD and MD-V3 are recommended by the suppliers for high dose range, although the sensitivity curves show that EBT3 and EBT4 could also be used for relatively high dose ranges. All films have orientation effects, with EBT3 the worst. An important characteristic of MD-V3 is that the LRA remains similar, irrespective of delivered dose. These comparative characteristics are intended to be informative for clinical practice involving Gafchromic film use in high dose therapy applications. Recommendations from this study are to use EBT4 for dosimetry in lower-dose applications, provided that both calibration and clinical timings post-irradiation are kept similar, while MD-V3 is the preferred film for high-dose procedures.</p>","PeriodicalId":48490,"journal":{"name":"Physical and Engineering Sciences in Medicine","volume":" ","pages":"1425-1437"},"PeriodicalIF":2.0,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144627480","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Maxim QA efficiency and accuracy.","authors":"Dean Wallace, Mikel Byrne, Kelvin Hiscoke, Trent Aland","doi":"10.1007/s13246-025-01589-z","DOIUrl":"10.1007/s13246-025-01589-z","url":null,"abstract":"<p><p>There are several commercial quality assurance (QA) software solutions for use in radiotherapy, but currently none are fully integrated with the existing c-arm linear accelerators. MaximQA (Varian Medical Systems, Inc. Palo Alto, CA) is a solution that integrates with both the TrueBeam and Halcyon linacs to capture and analyse QA tasks automatically. Currently the software supports a limited number of tests for the Halcyon and TrueBeam, including CBCT for both linacs, dynamic multi-leaf collimator tests (DMLC) for TrueBeam, and the Winston-Lutz test for Halcyon. This study investigates the efficiency gains of an integrated QA system compared to non-integrated QA software and evaluates its accuracy against another commercial product, DoseLab (Varian Medical Systems, Inc., Palo Alto, CA). Efficiency was assessed through a timing study by measuring the time required for CBCT, Winston-Lutz, and DMLC QA tasks in MaximQA and DoseLab. Accuracy was evaluated by comparing analysis results for the same tests in both software packages. The timing study showed that the integrated system substantially reduced QA task duration. A single CBCT QA analysis was 1 to 3 min faster, while DMLC and Winston-Lutz tests each saved 3 to 5 min excluding any export of images required. Comparison of analysis results indicated similar outcomes for most parameters, though some variations arose due to differences in calculation methodologies. Overall, the use of an integrated QA program decreased the time required to undertake these tests while maintaining high accuracy when compared to a previously established QA product.</p>","PeriodicalId":48490,"journal":{"name":"Physical and Engineering Sciences in Medicine","volume":" ","pages":"1351-1357"},"PeriodicalIF":2.0,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144601964","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-centric clinical implementation of the remote and automated quality control programme for digital imaging in Malaysia: challenges and pitfalls.","authors":"Vinoshah S Ravichandran, Nur Ammi Hamzah, Li Kuo Tan, Virginia Tsapaki, Olivera Ciraj Bjelac, Noramaliza Mohd Noor, Nur Syahiirah Mohamad Mokhtar, Nur Hasyimah Abd Rashid, Adiela Saiful Fazad, Nur Hafizah Zakaria, Siti Norsyafiqah Mohd Mystafa, Wan Nur Ain Wan Ghazali, Nur Shahidatul Akma Mohd Yusoff, Norafatin Khalid, Chai Hong Yeong, Muhammad Khalis Abdul Karim, Jeannie Hsiu Ding Wong","doi":"10.1007/s13246-025-01590-6","DOIUrl":"10.1007/s13246-025-01590-6","url":null,"abstract":"<p><p>The International Atomic Energy Agency (IAEA) has developed a methodology for a remote and automated quality control (QC) programme for digital radiography (DR) units. The purpose of this paper is to report the results of the implementation of the methodology in four hospitals in Malaysia. The IAEA methodology provides multiple image quality metrics by using dedicated software and standard, easily available materials to construct phantoms at a reasonably low cost. Nine QC phantoms were constructed and distributed across these institutions, with data collected daily or weekly analysed using the Python implementation of the Automated Tool for Image Analysis software. Image quality metrics, including signal-difference-to-noise ratio (SDNR), signal-noise ratio (SNR), modulation transfer function (MTF) and detectability index (d') were assessed on 11 digital radiography units from four different manufacturers. The use of diverse imaging protocols resulted in statistically significant differences in all the image quality metrics across the different units. For the processed image protocols, the median SDNR values ranged (12.2-17.5) and (9.1-17.9), respectively and were less affected by the protocol variations compared to SNR values. The d' 0.3 mm ranged (4.8-7.1) and (3.4-6.2), while the d' 4 mm variation ranged (73-115) and (83-130), respectively. The MTF values were strongly correlated between the horizontal and vertical MTFs, as well as between the MTF levels at 10%, 20% and 50%. Across different DR units, there were significant differences in the image quality metrics, mainly due to the different acquisition protocols employed. Clinical protocols have inherent image post-processing that can significantly alter the image quality metrics values compared to the raw image. The IAEA methodology is a useful tool to track the performance of DR units over time. Recommendations for the wider implementation of this methodology would include standardising the acquisition protocol by means of setting a specific QC protocol template to ensure consistency in the image acquisition.</p>","PeriodicalId":48490,"journal":{"name":"Physical and Engineering Sciences in Medicine","volume":" ","pages":"1359-1374"},"PeriodicalIF":2.0,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144660831","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comment on \"Subclinical tremor differentiation using long short-term memory networks\".","authors":"Hinpetch Daungsupawong, Viroj Wiwanitkit","doi":"10.1007/s13246-025-01559-5","DOIUrl":"10.1007/s13246-025-01559-5","url":null,"abstract":"","PeriodicalId":48490,"journal":{"name":"Physical and Engineering Sciences in Medicine","volume":" ","pages":"1485"},"PeriodicalIF":2.0,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144692084","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A distributed adaptive network framework for ERP-Based classification of multichannel EEG signals.","authors":"Fatemeh Afkhaminia, Mohammad Bagher Shamsollahi, Tahereh Bahraini","doi":"10.1007/s13246-025-01578-2","DOIUrl":"10.1007/s13246-025-01578-2","url":null,"abstract":"<p><p>Understanding brain function is one of the most challenging areas in brain signal processing. This study introduces a novel framework for electroencephalography (EEG) signal classification based on distributed adaptive networks using diffusion strategy. Our approach models the brain as a multitask network, where EEG electrodes are considered as nodes of this network. The network parameters are dynamically optimized based on the data from the nodes and inter-node cooperation. The proposed framework, which comprises network modeling and diffusion-based adaptation using the adapt then combine (ATC) algorithm, has been validated on different types of data. Experimental results indicate that the proposed framework outperforms common methods in classifying EEG data based on event-related potential (ERP) pattern identification, particularly in scenarios where machine learning-based models struggle with limited data. Furthermore, its ability to adapt to the non-stationary and dynamic nature of EEG signals and its efficient real-time implementation make this approach ideal for brain-computer interface (BCI), cognitive neuroscience, and clinical applications.</p>","PeriodicalId":48490,"journal":{"name":"Physical and Engineering Sciences in Medicine","volume":" ","pages":"1207-1224"},"PeriodicalIF":2.0,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144692082","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}