Medical & Biological Engineering & Computing最新文献

{"title":"Comparison of raw-signal, feature-based and hybrid representations for machine-learning classification of retinal diseases from pattern electroretinography.","authors":"Abdullah Bilal Aygün, Rumeysa Açıkgöz, Mehmet Coşkun, Mehmet Akif Erden","doi":"10.1007/s11517-026-03698-7","DOIUrl":"https://doi.org/10.1007/s11517-026-03698-7","url":null,"abstract":"<p><p>Pattern electroretinography (PERG) is a non-invasive test of macular and retinal ganglion cell function, but its low amplitude and inter-subject variability limit its diagnostic use. This study compares three representations of the transient PERG waveform for the detection of six retinal and optic nerve conditions - Birdshot chorioretinopathy, cone-rod dystrophy, hereditary optic atrophy, macular dystrophy, retinitis pigmentosa and Stargardt disease - against healthy controls, using the publicly available PERG-IOBA dataset. A total of 672 recordings from 336 examinations of 304 participants were preprocessed in six stages, and three representations were evaluated under one identical protocol: the raw 255-point amplitude-preserving waveform; 60 engineered clinical, spectral, wavelet and complexity descriptors; and a hybrid representation of 343 variables combining both, with multi-wavelet features from db4, sym2 and coif3. Performance was estimated by 10 × 5-fold participant-grouped cross-validation, in which both eyes and all repeat visits of an individual were kept in the same partition, with nested hyperparameter selection and all data-derived preprocessing fitted inside each training fold. Under the pre-specified primary classifier, the raw waveform gave the highest pooled out-of-fold AUC in four of the six classes (0.906 in cone-rod dystrophy, 0.842 in retinitis pigmentosa, 0.755 in macular dystrophy, 0.673 in hereditary optic atrophy) and the engineered descriptors in the remaining two (0.803 in Stargardt disease, 0.657 in Birdshot chorioretinopathy). The hybrid representation was highest in none, falling below the raw waveform in five of six classes by 0.002 to 0.018, and no non-linear classifier showed a hybrid advantage above 0.006 in mean AUC. A sensitivity analysis identified what does matter: removing the absolute amplitude scale by per-signal z-scoring cost the raw representation 0.055 AUC on average but the engineered descriptors only 0.023, and reversed the ordering of the three representations. Engineered descriptors did not improve on the waveform, but they made the problem tractable for linear models, which gained 0.128 in mean AUC from the reduction in dimensionality. These internally validated findings indicate that absolute PERG amplitude should be preserved rather than normalised away; participant-independent, multi-centre validation is required before clinical application.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":""},"PeriodicalIF":3.1,"publicationDate":"2026-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149225206","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":"Visual estimation of guidewire contact forces by integrating multi-scale features and region mask.","authors":"Jianming Guo, Dong Yang, Nan Xiao, Jiaqian Li, Qian Zheng, Wei Wei","doi":"10.1007/s11517-026-03690-1","DOIUrl":"https://doi.org/10.1007/s11517-026-03690-1","url":null,"abstract":"<p><p>During percutaneous coronary intervention (PCI), the contact force between the guidewire and the vessel wall is a key mechanical factor in vessel injury, yet it cannot be measured directly with conventional imaging. Although sensor-based methods measure the contact force directly, they increase both the complexity and the cost of the instrumentation. A sensorless guidewire contact force estimation method is therefore proposed, in which multi-scale feature extraction is combined with region mask mean pooling to learn an end-to-end mapping from conventional planar images to the contact force in each of the four predefined contact regions. The proposed method is evaluated on a dataset of 343 samples acquired on a 3D-printed vessel phantom, achieving a mean absolute error (MAE) of 0.861 g and a root mean square error (RMSE) of 1.036 g. Ablation experiments show that the region mask reduces the MAE by 79% relative to the identical architecture without the region mask. Requiring neither additional sensing hardware nor complex physical modeling, the method estimates the contact force from visual information alone. These results support the feasibility of vision-based sensorless contact force estimation and suggest a possible route toward force feedback in interventional procedures.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":""},"PeriodicalIF":3.1,"publicationDate":"2026-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149225220","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":"Single-Lead EEG-based virtual reality motion sickness detection using interpretable feature selection.","authors":"Min Zhu, Chengcheng Hua, Qiuyu Wang, Jingqi Fu, Sheng Xiang, Zhong Wei, Jia Liu","doi":"10.1007/s11517-026-03702-0","DOIUrl":"https://doi.org/10.1007/s11517-026-03702-0","url":null,"abstract":"<p><strong>Background: </strong>Electroencephalography (EEG) is a key method for detecting virtual reality motion sickness (VRMS), but multi-lead recordings and high-dimensional features increase processing complexity, device cost, and user burden.</p><p><strong>Objective: </strong>This study proposes a SHapley Additive exPlanations (SHAP)-based single-lead modeling and feature-screening framework to reduce the number of EEG leads and feature dimensions for lightweight VRMS detection.</p><p><strong>Method: </strong>Six classical deep-learning EEG decoding models were used to assess the contribution of each lead via SHAP values and identify the optimal single lead. Handcrafted features were then extracted and selected using a correlation-SHAP strategy, and XGBoost was adopted for final classification.</p><p><strong>Results: </strong>FT11 showed competitive and relatively consistent performance across the evaluated models and was selected as the target lead after integrating SHAP-based lead importance, single-lead classification performance, frequency-band differences, and physiological relevance. Under the window-level tenfold cross-validation protocol used in this study, the proposed model achieved an accuracy of 87.53% ± 0.73% using seven selected features, which was 9.55 percentage points lower than the full-lead full-feature model and 0.69 percentage points lower than the single-lead full-feature model.</p><p><strong>Significance: </strong>These findings demonstrate an interpretable framework for identifying a compact single-lead and feature configuration for VRMS-related EEG discrimination. An additional post hoc LOSO test of the globally selected configuration showed limited performance on held-out participants. Because the lead, classifier, and feature configuration had been selected globally before LOSO testing, these results should not be interpreted as a fully independent estimate of cross-subject generalization.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":""},"PeriodicalIF":3.1,"publicationDate":"2026-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149225170","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":"Design and optimization of a multi-band spiral antenna for frequency-dependent hyperthermia heating in head and neck cancer treatment.","authors":"Nitika Sharma, Rajesh Khanna, Mayank Kumar Rai","doi":"10.1007/s11517-026-03701-1","DOIUrl":"https://doi.org/10.1007/s11517-026-03701-1","url":null,"abstract":"<p><p>This research presents a compact multi-band spiral antenna designed for frequency-dependent hyperthermia therapy in head and neck cancers. The antenna is designed on a low-cost FR4 substrate with overall dimensions of 70 × 70 × 1.6 mm³ and is optimized to cover the 915 MHz, 1.8 GHz, and 2.45 GHz frequency bands. Its geometry enables stable tri-band resonance with excellent impedance matching, ensuring efficient energy transfer into biological tissues. The antenna's performance is evaluated through S-parameter analysis, specific absorption rate (SAR) distribution, penetration depth (PD), and effective field size (EFS) using homogeneous and heterogeneous phantoms, as well as the Katja voxel model with and without a tumor. The thermal simulations confirm that therapeutic temperatures of 41-45 °C are achieved with low input powers between 2.5-6 W. At 2.45 GHz, the antenna provides localized heating for superficial tumors (≤ 3 cm), at 1.8 GHz it is effective for intermediate tumor depths (3-3.6 cm), and at 915 MHz it achieves deeper penetration for tumors located beyond 3.6 cm. Experimental validation using head-simulating liquid and ex vivo chicken tissue demonstrates good agreement with simulation results, confirming reliable and efficient heating. The proposed applicator offers a practical and versatile solution, enabling personalized hyperthermia therapy across a wide range of tumor depths in head and neck cancer treatment.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":""},"PeriodicalIF":3.1,"publicationDate":"2026-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149217214","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":"UC-DiffuMambaNet: dual-scale state propagation and window attention for ordinal Mayo endoscopic grading.","authors":"Pan Ma, Hao Zhou, Hao Wu, Yi Li, Guowu Yuan","doi":"10.1007/s11517-026-03694-x","DOIUrl":"https://doi.org/10.1007/s11517-026-03694-x","url":null,"abstract":"<p><p>Automated four-grade Mayo Endoscopic Score (MES) classification is challenging because inflammatory severity depends not only on subtle local mucosal changes, but also on the collective spatial pattern of abnormalities that may be separated across the visible field. We present UC-DiffuMambaNet, a hybrid architecture that integrates dual-scale severity-evidence propagation, two-dimensional contextual coordination, and hierarchical ordinal prediction. The Severity Mamba Mixer first strengthens informative mucosal features through locally gated multi-scale enhancement and then processes them with parallel short- and long-retention selective state-space branches. This design preserves local feature continuity while linking separated severity cues over complementary contextual ranges. Medical Window Attention subsequently coordinates fine intra-window structure with communication across windows, complementing state propagation with explicit two-dimensional spatial context. Finally, an Auxiliary Hierarchical Classification Head refines ordered MES assessment through feedback-guided hierarchical prediction. Across three independent seeds, UC-DiffuMambaNet achieved ACC/QWK/Macro-F1 of 91.26%/96.28%/90.70% on Dataset A and 79.95%/87.65%/74.17% on Dataset B (LIMUC). Compared with the architecture-matched MambaVision-T baseline, ACC increased by 3.01 and 2.90 percentage points, respectively. Factorial and internal ablations confirmed complementary contributions from the three modules. Together, these components provide an integrated framework for ordinal MES assessment.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":""},"PeriodicalIF":3.1,"publicationDate":"2026-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149217161","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":"Correction: multi-source information fusion using CNN-LSTM-attention for bone layer recognition in robotic orthopedic grinding.","authors":"Kai Yang, Qingxuan Jia, Juxiang Huang, Gang Chen, Chao Feng, Yunfeng Xu","doi":"10.1007/s11517-026-03696-9","DOIUrl":"https://doi.org/10.1007/s11517-026-03696-9","url":null,"abstract":"","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":""},"PeriodicalIF":3.1,"publicationDate":"2026-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149197416","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":"Implementation of the first practical automatic micropulse blood pressure measurement device using the published revolutionary Newcastle technique.","authors":"Alan Murray, Dingchang Zheng, Clive J Griffiths, Chengyu Liu, David J Graham, Jeffrey Neasham, Adrian J Cossor","doi":"10.1007/s11517-026-03687-w","DOIUrl":"https://doi.org/10.1007/s11517-026-03687-w","url":null,"abstract":"<p><p>True accuracy for blood pressure measurement devices is required, but current techniques only estimate, deriving measurements from various private sets of clinical measurements. We previously reported a new technique that detects micropulse arterial opening, present only for beats between systole and diastole, enabling direct determination of systolic and diastolic blood pressures (SBP & DBP). We have now developed a first automated device using our technique and have compared its accuracy with current gold standard auscultatory measurements in 32 subjects. The differences between our novel device and simultaneous auscultatory measurements were for SBP -2.5 ± 3.1 mmHg, and for DBP -1.0 ± 1.2 mmHg (mean ± standard deviation SD). Results were well within the clinical international standard's limits of 5 ± 8 mmHg.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":""},"PeriodicalIF":3.1,"publicationDate":"2026-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149172093","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":"RDB CycleGAN-CBAM: an integrated deep learning model for low-dose ct image denoising.","authors":"Osman Assaf, Albert Güveniş","doi":"10.1007/s11517-026-03689-8","DOIUrl":"https://doi.org/10.1007/s11517-026-03689-8","url":null,"abstract":"<p><p>Computed Tomography (CT) is one of the largest contributors to radiation exposure from medical imaging, which can induce DNA damage and increase cancer risk. Reducing CT radiation dose to improve patient safety inherently increases image noise and artifacts. Generative adversarial networks (GANs) have shown promise for unsupervised low-dose CT (LDCT) denoising. Building on this, RDBCycleGAN-CBAM, a CycleGAN-based model that integrates residual dense blocks (RDBs) and convolutional block attention modules (CBAM), was developed to effectively denoise quarter-dose CT images while preserving structural detail. The model was trained on unpaired quarter-dose and full-dose CT scans from the NIH-AAPM-Mayo dataset using adversarial (LSGAN), cycle-consistency, and identity losses. Evaluation on held-out test slices was performed using PSNR and SSIM as the primary image-quality metrics. The results demonstrate that the proposed RDBCycleGAN-CBAM method demonstrated improvements that were competitive with values reported in previous studies, achieving mean improvements of + 3.973 dB in PSNR and + 0.053 in SSIM relative to quarter-dose inputs. Shapiro-Wilk tests for PSNR and SSIM motivated the use of the nonparametric Wilcoxon signed-rank test, by which highly significant improvements across both metrics (PSNR and SSIM) were demonstrated. The very large rank-biserial correlation values (1.0) indicate that nearly all test images experienced substantial quality improvement. Furthermore, the narrow bootstrap confidence intervals for the mean differences suggest that these improvements are consistent across the dataset. These findings support further investigation of RDBCycleGAN-CBAM as a vendor-neutral image-domain denoising approach, while external validation and reader studies are required to establish its clinical utility and potential for dose reduction.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":""},"PeriodicalIF":3.1,"publicationDate":"2026-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149167767","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":"Phantom system for endotracheal tube cuff seal testing during mechanical ventilation.","authors":"Delu Zheng, Tamaralayefa B Agbiki, Ricardo Correia, Chenyang He, Sergiy Korposh, Sandor Erdody, Barrie R Hayes-Gill, David W Hewson, Andrew M Norris, Stephen P Morgan","doi":"10.1007/s11517-026-03691-0","DOIUrl":"https://doi.org/10.1007/s11517-026-03691-0","url":null,"abstract":"<p><p>Leakage of orogastric secretions past endotracheal tube (ETT) cuffs contributes to ventilator-associated pneumonia (VAP), yet in vivo tracheal seal testing is impractical, requiring phantom evaluation. As the rigid cylinder specified in EN ISO 5361:2023 does not adequately represent in vivo airway conditions, a benchtop cuff seal testing system was developed incorporating a tracheal model, lung model, mechanical ventilation, and automatic cuff pressure control. A rigid cylinder and a bio-inspired tracheal model with tissue-like mechanical properties were evaluated using five high-volume low-pressure (HVLP) polyvinyl chloride (PVC) cuffed ETTs and five polyurethane (PU) cuffed ETTs, with and without mechanical ventilation, and with and without automatic cuff inflation. For PVC cuffed ETTs, fluid leakage rate was highest in the cylinder phantom compared with the bio-inspired phantom, ~ 30% lower without mechanical ventilation and ~ 50% with mechanical ventilation (both p < 0.001). In contrast, PU ETTs exhibited minimal leakage. Long-term testing revealed intracuff pressure decay over 48 h at 37 °C for both PVC (12.24 ± 1.13 cmH2O) and PU (9.06 ± 1.00 cmH2O) cuffed ETTs, while automatic cuff pressure control further reduced leakage for PVC ETTs. The bio-inspired phantom revealed sealing behaviours not captured by cylinder testing, supporting clinically relevant evaluation of ETT sealing performance and cuff materials.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":""},"PeriodicalIF":3.1,"publicationDate":"2026-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149139039","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":"Cardiovascular function changes following lung resection: a computational model to compare afterload increase and contractility loss mechanisms.","authors":"Shiting Huang, Sanjay Pant, Sean McGinty, Richard Good, Ben Shelley, Ankush Aggarwal","doi":"10.1007/s11517-026-03682-1","DOIUrl":"https://doi.org/10.1007/s11517-026-03682-1","url":null,"abstract":"<p><p>Functional limitation after lung resection surgery has been consistently documented in clinical studies, and right ventricle (RV) dysfunction has been hypothesized as a contributing factor. However, the mechanisms of RV dysfunction after lung resection remain unclear, particularly whether change in afterload or contractility is the underlying cause. This study reviews the clinical literature on this topic and proposes a lumped parameter model to simulate the effects of lung resection. Our model allowed us to isolate the two mechanisms and compare them. Subsequently, the simulation outputs were compared against the literature. Our rigorous approach included local and global sensitivity analyses to evaluate the effect of model parameters, both individually and collectively. The two mechanisms produced the same trends for all outputs except three: RV systolic pressure (RVSP) and pulmonary artery systolic and diastolic pressure (PASP and PADP) changed in opposite directions. Furthermore, our model indicates that obtaining pressure measurements in the clinic may help quantify the precise combination of the two mechanisms. This study presents the first mathematical model of the effect of lung resection on the cardiovascular system. Further refinement, combined with additional clinical data, will help us predict RV dysfunction and pave the way towards improving outcomes for lung cancer patients.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":""},"PeriodicalIF":3.1,"publicationDate":"2026-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149138958","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}