IEEE Transactions on Biomedical Engineering最新文献

{"title":"MorphDiff-ECG: A Diffusion Model for Multimodal Feature Fusion of ECG Signals.","authors":"Lei Zhang, Cheng Su, Mengying Ou, Mengyue Wang, Zhichao Xu, Shiyu Wei, Chang Liu, Tao Lei, Yong Wan","doi":"10.1109/TBME.2026.3735103","DOIUrl":"10.1109/TBME.2026.3735103","url":null,"abstract":"<p><strong>Objective: </strong>Electrocardiogram (ECG) generation is crucial for computer-aided diagnosis and medical data augmentation, yet existing generative models struggle with waveform fidelity, morphological accuracy, and rhythm consistency.</p><p><strong>Methods: </strong>We propose a novel diffusion model for multimodal feature fusion of ECG signals, namely MorphDiff-ECG. Inspired by the hierarchical \"morphology-rhythm-trend\" logic that clinicians follow when interpreting ECGs, we designed a two-stage hierarchical decoding architecture comprising two core stages. The first stage, termed Feature Decoupling Enhancement (FDE), comprises three parallel branches dedicated to modeling the trend, periodic, and morphological components of physiological signals. This explicit decoupling design not only aligns the learned features with clinically interpretable terms, but also enhances the model's sensitivity to abnormal morphological patterns. Specifically, the trend branch captures low-frequency baseline drift and global amplitude variations, ensuring stable signal horizons; the periodic branch extracts rhythmic cardiac patterns through multi-scale time-frequency analysis, preserving both regular and irregular heart rate dynamics; the morphological branch isolates fine-grained local waveform geometry, enabling precise reconstruction of diagnostically critical features such as P-waves, QRS complexes, and T-waves. This tripartite decomposition embeds the hierarchical diagnostic logic of cardiology directly into the network architecture, providing both physiological plausibility and mechanistic interpretability. The second stage, termed Cross-Feature Attention Fusion (CFAF), integrates the decoupled features through an efficient attention-based mechanism to reconstruct complete physiological signals. To further preserve critical high-frequency transients, exemplified by QRS complexes, we introduce a hybrid training objective that combines time-domain reconstruction with wavelet-based time-frequency loss. This approach takes advantage of the inherent time-frequency coupling of the ECG signals, ensuring structural integrity and spectral fidelity in the generated outputs.</p><p><strong>Results: </strong>Experiments on three classic public datasets show that MorphDiff-ECG outperforms state-of-the-art methods in generation fidelity on the MIT-BIH, CPSC2020, and extended nine-class CPSC2018 benchmarks. Compared to the best-performing baseline, it reduces PRD by 1.97%, FD by 2.38%, RMSE by 10.39% and MAE drops by 0.61%.</p><p><strong>Conclusion: </strong>This work provides an effective solution for high-quality and controllable ECG synthesis.</p><p><strong>Significance: </strong>MorphDiff-ECG advances cardiac diagnosis and personalized medical data generation through improved quality of ECG synthesis.</p>","PeriodicalId":13245,"journal":{"name":"IEEE Transactions on Biomedical Engineering","volume":"PP ","pages":""},"PeriodicalIF":4.4,"publicationDate":"2026-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148989589","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"PointSenseNet: A Hierarchical Point-Cloud-Based Reconstruction Framework with Application to Localized Inverse Electrocardiographic Mapping.","authors":"Yunchao Wu, Hiroshi Seno, Yilin Wang, Yuka Seki, Masatoshi Yamazaki, Ichiro Sakuma, Naoki Tomii","doi":"10.1109/TBME.2026.3734484","DOIUrl":"10.1109/TBME.2026.3734484","url":null,"abstract":"<p><strong>Objective: </strong>Cardiac excitation mapping based on intracardiac ECG is essential for identifying arrhythmogenic substrates in persistent atrial fibrillation, yet existing methods are limited by spatial resolution and the requirement of electrode-tissue contact. This study aims to develop a high-resolution localized non-contact cardiac mapping method.</p><p><strong>Methods: </strong>We propose a hierarchical point-cloud deep learning model, PointSenseNet (PSN), designed to address sparse signal reconstruction problems exemplified by ECG mapping. Using a cardiac electrophysiology simulation model, we evaluated PSN across multiple non-contact distances and compared it with the conventional inverse ECG mapping method. In addition, we investigated the effectiveness of the virtual intermediate point cloud layer incorporated into PSN.</p><p><strong>Results: </strong>Across various excitation patterns, PSN demonstrated improved predictive accuracy and spatial resolution compared with the conventional method. At a non-contact distance of 10 mm, PSN successfully reconstructed critical structures such as conduction block regions and achieved a relative error (normalized membrane potential) below 10% and activation time error below 6 ms. The inclusion of the intermediate point cloud layer further improved performance across different non-contact distances.</p><p><strong>Conclusion: </strong>PSN demonstrates promising performance for non-contact ECG mapping, achieving higher predictive accuracy than that of current methods. Owing to its strong generalization capability, PSN also holds potential for broader application to sparse signal reconstruction tasks beyond ECG mapping.</p><p><strong>Significance: </strong>These findings have the potential to substantially reduce the procedural burden associated with contact-based mapping in future clinical applications while providing more precise mapping information that may facilitate the development of novel ablation strategies.</p>","PeriodicalId":13245,"journal":{"name":"IEEE Transactions on Biomedical Engineering","volume":"PP ","pages":""},"PeriodicalIF":4.4,"publicationDate":"2026-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148989593","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Low-gamma wavelet entropy as a contextual signature descriptor of the epileptogenic zone in peri-onset SEEG.","authors":"Valentin Ricordel, Alina Voda, Gildas Besancon, Philippe Kahane","doi":"10.1109/TBME.2026.3733850","DOIUrl":"10.1109/TBME.2026.3733850","url":null,"abstract":"<p><strong>Objective: </strong>In drug-resistant focal epilepsy, surgery - which requires accurate localization of the seizure origin - is the only curative treatment. When non-invasive exploration fails, stereo-electroencephalography (SEEG) provides the best a priori localization but clinical practice still relies on subjective and perfectible visual analysis.</p><p><strong>Method: </strong>This work investigates whether wavelet-based entropy features with contextual descriptors, from peri-onset of SEEG-recorded seizures can characterize the region to remove. For interpretability, a logistic regression classifier is trained.</p><p><strong>Results: </strong>Leave-One-Subject-Out cross-validation on a post-surgery seizure-free cohort shows good performance (sensitivity: 89% [95% CI: 68-97], specificity: 91% [95% CI: 87-95]) in identifying epileptogenic sensors. The model flags 14% of contacts, capturing 86% of the reference. Frequency content in the 40-80$,mathrm{mathbf{Hz}}$ band appears particularly informative.</p><p><strong>Conclusion: </strong>The proposed method enables accurate and interpretable identification of epileptogenic contacts, with potential for patient-level screening pending validation on larger datasets.</p><p><strong>Significance: </strong>An efficient, parsimonious, and explainable model can support more targeted surgical decision-making while reducing the SEEG review burden and improving outcomes.</p>","PeriodicalId":13245,"journal":{"name":"IEEE Transactions on Biomedical Engineering","volume":"PP ","pages":""},"PeriodicalIF":4.4,"publicationDate":"2026-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148989369","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Rhomboid Flap Simulations under Anisotropic Skin Tension Conditions: A Shape-based Validation Study through Porcine Experiments and Patient Photograph Analysis.","authors":"Wenzhangzhi Guo, Jon S Heiselman, Kyvia Pereira, Mark I Petrovic, Long Van Tran Ha, Joel C Davies, Eitan Grinspun, Michael I Miga, Lueder A Kahrs","doi":"10.1109/TBME.2026.3734486","DOIUrl":"10.1109/TBME.2026.3734486","url":null,"abstract":"<p><strong>Objective: </strong>Finite element simulation has the potential to assist facial reconstructive surgery planning by providing biomechanical insight into post-operative tissue behaviour. However, relatively few studies quantitatively evaluate the agreement between simulated and observed surgical outcomes. In this paper, we propose a pipeline for validating simulated suture lines using registered patient photograph pairs and porcine samples.</p><p><strong>Methods: </strong>To analyze patient photographs collected from the literature, facial landmarks are extracted from pre- and post-operative images. For images containing missing or occluded facial regions, Stable Diffusion inpainting is used to improve landmark detection accuracy. Additionally, 12 skin flaps are performed on two pigs and reconstructed as coloured 3D point clouds obtained using a stereo camera. The reconstructed point clouds are projected into 2D images. For both datasets, pre- and post-operative images are registered and compared with the corresponding simulation results, enabling quantitative evaluation of the simulated suture lines.</p><p><strong>Results: </strong>Overall, 18 of 20 analyzed patient photographs and 11 of 12 porcine samples exhibited a normalized Hausdorff distance at or below 8%, indicating close geometric agreement between the observed and simulated suture lines.</p><p><strong>Conclusion: </strong>Quantitative comparison with clinical and experimental data demonstrates the feasibility of assessing agreement between simulated and observed reconstructive surgery outcomes.</p><p><strong>Significance: </strong>These results demonstrate that finite element simulations can reproduce observed suture line patterns and suggest that relaxed skin tension lines (RSTLs) play an important role in accurately simulating reconstructive surgery procedures.</p>","PeriodicalId":13245,"journal":{"name":"IEEE Transactions on Biomedical Engineering","volume":"PP ","pages":""},"PeriodicalIF":4.4,"publicationDate":"2026-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148989757","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"FreqDINO++: A Frequency-Guided Multi-Task Routing Vision Foundation Model for Universal Ultrasound Analysis.","authors":"Qing Xu, Yixuan Zhang, Yue Li, Xiangjian He, Qian Zhang, Mainul Haque, Rong Qu, Wenting Duan, Jieyun Bai, Zhen Chen","doi":"10.1109/TBME.2026.3733527","DOIUrl":"10.1109/TBME.2026.3733527","url":null,"abstract":"<p><p>Ultrasound image analysis plays a crucial role in cancer screening and prenatal diagnosis, yet comprehensive assessment requires jointly addressing tasks such as lesion segmentation and benign-malignant classification. While recent vision foundation models have shown remarkable universal representations, unlocking their potential for ultrasound is bottlenecked by the considerable domain gap from natural images. Existing methods typically fine-tune heavy vision encoders for isolated tasks, incurring substantial computational overhead while overlooking the underlying commonalities across heterogeneous tasks. In this work, we propose FreqDINO++, a frequency-guided multi-task routing vision foundation model for universal ultrasound analysis. We first introduce a Multi-task Routing Adapter (MR-Adapter) to support parameter-efficient integration of task-common and task-specific knowledge, a Frequency-aware Feature Enhancer (F$^{2}$-Enhancer) is then designed to capture the rich multi-scale frequency characteristics of ultrasound images, and a Task-aligned Collaborative Decoder (TC-Decoder) is devised to promote collaboration between dense and global prediction tasks through global-local token interaction. Extensive experiments on large-scale multi-task and external single-task ultrasound benchmarks demonstrate that FreqDINO++ consistently outperforms strong baselines and recent foundation models across 27 diverse clinical task scenarios, while also showing promising generalization to unseen data.</p>","PeriodicalId":13245,"journal":{"name":"IEEE Transactions on Biomedical Engineering","volume":"PP ","pages":""},"PeriodicalIF":4.4,"publicationDate":"2026-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148976727","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Dual-Curriculum and Orthogonal Domain Prototype Learning for Domain Generalization Fundus Image Segmentation.","authors":"Cangxin Li, Shichen Liao, Haoyu Chen, Tao Peng, Fei Shi, Weifang Zhu, Xinjian Chen, Dehui Xiang","doi":"10.1109/TBME.2026.3730099","DOIUrl":"10.1109/TBME.2026.3730099","url":null,"abstract":"<p><strong>Objective: </strong>Image segmentation plays a crucial role in retinal disease analysis and computer-aided clinical diagnosis. However, substantial appearance variations across imaging devices, clinical centers, and acquisition conditions often lead to significant performance degradation when segmentation models are deployed to previously unseen domains due to the distribution gap caused by cross-device discrepancies.</p><p><strong>Methods: </strong>A dual-curriculum learning strategy is proposed to handle diverse patterns and mitigate the domain shift between multi-source domain and target domain. Furthermore, a cross-level correlation consistency feature fusion strategy is proposed. This strategy captures the pixel-level correlations from low-level feature and maintains these correlations within high-level features. In addition, orthogonal domain prototypes of multi-source domains are constructed to better adapt to the multi-source domain generalization task.</p><p><strong>Results: </strong>Experiments on two multi-domain OCT image datasets demonstrate that the proposed method consistently outperforms state-of-the-art approaches. The results validate its effectiveness under cross-domain scenarios. Additional experiments on a color photography image dataset further confirm the robustness and generalizability of the proposed framework across diverse fundus imaging modalities.</p><p><strong>Conclusion: </strong>Our model trained on labeled fundus images from various manufacturers' devices achieves superior domain generalization when tested on fundus images from previously unseen devices.</p><p><strong>Significance: </strong>The proposed method is suitable for routine clinical use. By mitigating the need for device-specific annotations, the proposed method holds promise for routine clinical deployment and reducing manual labeling costs.</p>","PeriodicalId":13245,"journal":{"name":"IEEE Transactions on Biomedical Engineering","volume":"PP ","pages":""},"PeriodicalIF":4.4,"publicationDate":"2026-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149038369","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Spatial-Sequence Joint Modeling for Accelerated Multi-Model Diffusion MRI Microstructure Estimation.","authors":"Taohui Xiao, Cheng Li, Shoujun Yu, Wenxin Fan, Ruoyou Wu, Enqing Dong, Shanshan Wang","doi":"10.1109/TBME.2026.3733593","DOIUrl":"10.1109/TBME.2026.3733593","url":null,"abstract":"<p><strong>Objective: </strong>This study aims to develop an efficient and high-fidelity deep learning framework for accelerated multi-model diffusion MRI microstructure estimation using sparsely sampled q-space data.</p><p><strong>Method: </strong>We propose a shared-encoder, distinct-decoder framework. A dual-branch architecture within the shared encoder integrates convolutional neural network (CNN)-based spatial modeling for local tissue structures and Mamba-based sequence modeling for global structural priors. The fused features form a shared latent representation that regularizes the highly subsampled estimation process. Distinct decoders extract model-specific features from this latent space, enabling diverse parameter estimation across biophysical models. A joint estimation loss with tunable weights balances the multi-model learning objectives.</p><p><strong>Results: </strong>Extensive experiments on the Human Connectome Project (HCP), Alzheimer's Disease Neuroimaging Initiative (ADNI), and Tiantan clinical dataset demonstrate consistent improvements over six state-of-the-art methods under multiple q-space subsampling settings. The proposed framework supports 4.5×-27× q-space acceleration while achieving 3×-9× faster inference. Additional experiments further support the robustness and cross-dataset adaptability of the proposed framework.</p><p><strong>Conclusion and significance: </strong>The proposed framework enables accurate and reliable joint estimation of multi-model dMRI microstructural parameters from accelerated acquisitions and provides a practical solution for fast microstructural imaging.</p>","PeriodicalId":13245,"journal":{"name":"IEEE Transactions on Biomedical Engineering","volume":"PP ","pages":""},"PeriodicalIF":4.4,"publicationDate":"2026-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148944885","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Bound-Constrained Sparse Representation for Electrical Impedance Tomography.","authors":"Chun Zhang, Dong Liu","doi":"10.1109/TBME.2026.3732327","DOIUrl":"10.1109/TBME.2026.3732327","url":null,"abstract":"<p><strong>Objective: </strong>To develop a stable and physically constrained framework for absolute electrical impedance tomography (EIT) reconstruction.</p><p><strong>Methods: </strong>A bound constrained sparse representation (BC-SR) framework is proposed to embed structural and physical priors directly into the conductivity parameterization. A truncated graph Laplacian basis provides a low-dimensional representation, while a nonlinear bound-preserving mapping enforces prescribed conductivity ranges. The resulting latent variables are reconstructed using nonlinear optimization on 2D and 3D unstructured meshes.</p><p><strong>Results: </strong>Experiments involving ablation and noise-robustness studies, 2D, 2.5D, and 3D numerical simulations, and tank measurements show that BC-SR suppresses unstable artifacts and improves structural fidelity and physical consistency compared with conventional reconstruction methods.</p><p><strong>Conclusion: </strong>BC-SR improves the stability of absolute EIT reconstruction by constraining both the effective solution space and the admissible conductivity values.</p><p><strong>Significance: </strong>BC-SR enables the effective reconstruction dimension to be selected independently of the FEM mesh resolution and reduces dependence on case-specific regularization-weight tuning, providing a broadly applicable framework for 2D and 3D EIT on unstructured meshes.</p>","PeriodicalId":13245,"journal":{"name":"IEEE Transactions on Biomedical Engineering","volume":"PP ","pages":""},"PeriodicalIF":4.4,"publicationDate":"2026-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148919355","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Twisted String Actuators in Wearable Robotics: Review and Perspective across Rehabilitation, Assistance, Augmentation, and Haptics.","authors":"Callista Shekar Ayu Supriyono, Mihai Dragusanu, Monica Malvezzi","doi":"10.1109/TBME.2026.3732502","DOIUrl":"10.1109/TBME.2026.3732502","url":null,"abstract":"<p><p>Twisted String Actuators (TSAs) have emerged as a compelling soft actuation paradigm for wearable robotic systems, offering high force output within a compact, lightweight, and mechanically compliant form factor. This systematic review provides a comprehensive analysis of TSA-based wearable devices reported in the literature between 2013 and 2026, synthesizing 34 functional prototypes identified through a structured search of Scopus and Google Scholar databases. Devices are categorized according to two orthogonal frameworks: anatomical target segment and primary application domain. The review examines the biomimetic foundations of TSA operation, its analogy to the human musculoskeletal system, and the key mechanical design decisions governing actuator configuration, force transmission strategy, degrees of freedom, structural substrate, and sensing modality. A progressive shift from rigid toward soft and hybrid architectures is identified across the review period, consistent with user acceptance requirements in clinical and industrial settings. Intent-driven control strategies fusing surface electromyography and inertial measurement units have emerged as the dominant paradigm for assistive and augmentation devices. Persistent open challenges include the non-linear kinematics of TSA force-displacement relationships, the near-total absence of long-term mechanical durability data, and the lack of real-world clinical validation. Notable anatomical gaps, particularly at the shoulder and cervical complex, are identified as high-priority directions for future development. Collectively, the findings position TSA-based wearables as a rapidly maturing technology with strong potential for translation into mainstream rehabilitation and human augmentation platforms.</p>","PeriodicalId":13245,"journal":{"name":"IEEE Transactions on Biomedical Engineering","volume":"PP ","pages":""},"PeriodicalIF":4.4,"publicationDate":"2026-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148919417","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An Automated Framework for ML-Assisted Lung-Function Assessment From Raw EIT Recordings.","authors":"Anzhelika Mezina, Stepan Miklanek, Samuel Genzor, Jan Mizera, Vojtech Myska, Radim Burget","doi":"10.1109/TBME.2026.3724766","DOIUrl":"10.1109/TBME.2026.3724766","url":null,"abstract":"<p><strong>Objective: </strong>Spirometry indices such as FEV1 and the Tiffeneau index (TI) are the clinical gold standard for measuring airflow limitation but provide only global lung measurements without capturing regional ventilation variability. This study aims to develop an automated, machine learning (ML)-assisted pipeline that estimates spirometry-related lung-function indices and discriminates respiratory disease classes directly from electrical impedance tomography (EIT) signals, a radiation-free bedside modality with high temporal resolution.</p><p><strong>Methods: </strong>We processed raw EIT recordings from a Dräger Pulmovista device through an automated pipeline that reconstructs impedance curves, performs quality control, optionally applies adaptive denoising, and segments the forced vital capacity (FVC) maneuver into four phases (A-D) to enable robust feature extraction. We evaluated multiple regression and classification models with hyperparameter optimization across three tasks: FEV1 prediction, TI prediction, and multiclass respiratory disease classification.</p><p><strong>Results: </strong>AdaBoost achieved the best performance for FEV1 regression (mean squared error [MSE] = 0.1026, $R^{2}$ = 0.9096), while XGBoost performed best for TI regression (MSE = 0.0037, $R^{2}$ = 0.7363). A multilayer perceptron (MLP) achieved the highest disease classification performance (accuracy = 0.8947, balanced accuracy = 0.8857, F1-score = 0.8857, ROC-AUC = 0.9444).</p><p><strong>Conclusion: </strong>These findings demonstrate that automated maneuver segmentation combined with rigorous quality control and ML-based feature engineering enables reliable estimation of spirometry-derived indices and accurate discrimination of respiratory disease classes from EIT-derived descriptors alone.</p><p><strong>Significance: </strong>This pipeline establishes EIT as a viable radiation-free, bedside alternative for continuous, spirometry-informed pulmonary function monitoring, with potential to improve respiratory disease surveillance in settings where conventional spirometry is impractical.</p>","PeriodicalId":13245,"journal":{"name":"IEEE Transactions on Biomedical Engineering","volume":"PP ","pages":""},"PeriodicalIF":4.4,"publicationDate":"2026-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148911518","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}