IEEE Transactions on Biomedical Engineering最新文献

{"title":"FedSAM-3D: Adapter-Constrained Federated Adaptation for Transferable Medical Segmentation Foundation Models.","authors":"Xinran Wu, Rencheng Zheng, Yuxiang Dai, Hui Zhang, Xueqin Xia, Yu Cheng, Chengyan Wang, He Wang","doi":"10.1109/TBME.2026.3730634","DOIUrl":"https://doi.org/10.1109/TBME.2026.3730634","url":null,"abstract":"<p><strong>Objective: </strong>Transferring large-scale medical foundation models to specific clinical tasks remains challenging, particularly in multi-center scenarios with heterogeneous data distributions and privacy constraints. Existing adaptation strategies provide limited solutions for collaboratively adapting foundation models across institutions while preserving their transferable representations.</p><p><strong>Methods: </strong>We propose FedSAM-3D, a foundation model adaptation framework for multi-center medical image segmentation. Built upon the SAM-Med3D backbone, FedSAM-3D defines the collaborative optimization space within adapter parameters while keeping the pretrained backbone unchanged. Through federated optimization within this constrained adaptation space, our framework enables efficient cross-center knowledge aggregation without exchanging full model parameters, while allowing each client to adapt the foundation model to local medical data distributions.</p><p><strong>Results: </strong>FedSAM-3D was evaluated on multi-center abdominal organ and brain tumor segmentation datasets under federated adaptation and zero-shot evaluation settings. Across both tasks and multiple clinical datasets, FedSAM-3D generally outperformed ablation variants and existing segmentation methods, demonstrating improved adaptation performance and robustness across heterogeneous medical data distributions. Moreover, FedSAM-3D achieved improved generalization on unseen external datasets, including cross-modality evaluation, highlighting its ability to enhance the transferability of medical foundation models.</p><p><strong>Conclusion: </strong>FedSAM-3D provides an effective paradigm for federated transfer of medical foundation models, achieving improved adaptation performance and generalization while avoiding direct sharing of raw medical data across institutions.</p><p><strong>Significance: </strong>FedSAM-3D provides a parameter-efficient approach for transferring medical foundation models across institutions without directly sharing raw data, facilitating their potential deployment in diverse clinical environments. Our code is available at https://github.com/huavhuahua/FedSAM-3D.</p>","PeriodicalId":13245,"journal":{"name":"IEEE Transactions on Biomedical Engineering","volume":"PP ","pages":""},"PeriodicalIF":4.4,"publicationDate":"2026-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148887364","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":"Enhancing Prosthetic Wrist Precision With Robust $mathcal {H}_infty$ Control and Timoshenko Beam Modelling.","authors":"Shifa Sulaiman, Francesco Schetter, Paolino De Risi, Mohammad Gohari, Fanny Ficuciello","doi":"10.1109/TBME.2026.3722458","DOIUrl":"https://doi.org/10.1109/TBME.2026.3722458","url":null,"abstract":"<p><strong>Objective: </strong>The creation and implementation of control strategies that govern the motions of prosthetic hands are essential for enhancing operational efficiency and user satisfaction. This paper introduces an H-infinity ($mathcal {H}_infty$) controller employing Timoshenko beam theory, specifically designed for a tendon-driven soft continuum wrist associated with a prosthetic hand known as 'PRISMA HAND II'. This combined approach focuses on achieving robust control that can effectively manage uncertainties and disturbances, ensuring that the prosthetic wrist responds accurately to the user's intentions.</p><p><strong>Methods: </strong>By employing the Timoshenko modeling approach, kinematic and dynamic models of the soft wrist are established, which are used in $mathcal {H}_infty$ controller to compute required tendon forces for achieving desired hand movements. These tendon forces are used for computing output deflections of the wrist.</p><p><strong>Results: </strong>The proposed controller is compared with other controllers to analyse performance of the proposed controller. The proposed $mathcal {H}_infty$ controller performed better compared to a Neural Network (NN) based adaptive controller in terms of Root Mean Square Error (RMSE) and steady state error values. Experimental evaluation demonstrated the controller's effectiveness in regulating wrist motions during real-time.</p><p><strong>Conclusion: </strong>The implementation of the control strategy is crucial for advancing the capabilities of soft continuum prosthetics, allowing for more natural and intuitive movements that align with the needs of the user with lower computational effort and better accuracy of motions.</p><p><strong>Significance: </strong>Our study contributes to the ongoing progress in adaptable prosthetic technologies, establishing a basis for extensive use in cost-effective healthcare solutions employing an efficient controller.</p>","PeriodicalId":13245,"journal":{"name":"IEEE Transactions on Biomedical Engineering","volume":"PP ","pages":""},"PeriodicalIF":4.4,"publicationDate":"2026-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148887353","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":"Identifiability of Spectral Graph Model Parameters in Clinical MEG: Implications for Biophysical Interpretation and SOZ Localization.","authors":"Himanshu Kumar, Guhan Seshadri N P, Navya Nair, Imad Najm, Richard Burgess, Andreas Alexopoulos, Hiroatsu Murakami, Balu Krishnan","doi":"10.1109/TBME.2026.3730069","DOIUrl":"https://doi.org/10.1109/TBME.2026.3730069","url":null,"abstract":"<p><strong>Objective: </strong>We evaluate the practical identifiability and clinical utility of local spectral graph model (SGM) parameters estimated from resting-state magnetoencephalography (MEG) in drug-resistant epilepsy.</p><p><strong>Methods: </strong>A coupled excitatory-inhibitory SGM was fitted to MEG power spectra across 159 brain regions in 20 patients with temporal lobe epilepsy who achieved seizure freedom following surgery. Identifiability was assessed via boundary saturation analysis and inter-parameter correlations across four frequency bands. Identifiable parameters were tested for seizure onset zone (SOZ) discrimination.</p><p><strong>Results: </strong>Model fit was excellent (mean $r = 0.976$) and significantly exceeded a $1/f^beta$ baseline ($p < 10^{-10}$). Gain parameters ($g_{ei}$, $g_{ii}$) were robustly estimable, whereas the excitatory time constant ($tau _{e}$) showed 74% boundary saturation in broadband fits, reduced to ${sim }3%$ when restricted to 1-50 Hz. SOZ regions exhibited elevated $g_{ei}$ (Cohen's $d = +0.67$, FDR-corrected $p = 0.024$) and reduced $g_{ii}$ ($d = -0.55$, $p = 0.003$), with a composite biomarker achieving 2.6-fold improvement over chance.</p><p><strong>Conclusion: </strong>Gain parameters are robustly identifiable from clinical MEG and capture excitatory-inhibitory imbalance in the SOZ, whereas time constants require band-limited fitting. These findings motivate an identifiability-aware framework in which only parameters demonstrably constrained by data are interpreted.</p><p><strong>Significance: </strong>This is the first systematic assessment of neural mass model parameter identifiability in clinical epilepsy MEG, establishing practical guidelines for biophysical parameter interpretation.</p>","PeriodicalId":13245,"journal":{"name":"IEEE Transactions on Biomedical Engineering","volume":"PP ","pages":""},"PeriodicalIF":4.4,"publicationDate":"2026-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148880188","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":"Human Recorded Signal Properties of Endovascular EEG Compared to Conventional Scalp EEG.","authors":"Luis Torres Mailleux, Kota Araki, Modar Hassan, Yosuke Masuda, Hiroki Ishida, Hisayuki Hosoo, Aiki Marushima, Yuji Matsumaru, Kenji Suzuki","doi":"10.1109/TBME.2026.3730012","DOIUrl":"https://doi.org/10.1109/TBME.2026.3730012","url":null,"abstract":"<p><strong>Objective: </strong>Endovascular EEG (eEEG) has emerged as a brain monitoring technique that offers a balance between signal fidelity and invasiveness. Endovascular electrodes match subdural recordings in bandwidth and signal-to-noise ratio in animal studies, however, their signal properties remain sparsely quantified in humans. This study evaluated eEEG signals from five human participants undergoing intracarotid amobarbital injection (Wada test), while simultaneous scalp and endovascular EEG were recorded.</p><p><strong>Methods: </strong>All signals were preprocessed with artifact rejection and independent component analysis (ICA). Power spectral density (PSD), imaginary coherence (ImagC), phase-locking value (PLV), and amplitude envelope correlation (AmpC) were computed to quantify signal quality and functional connectivity.</p><p><strong>Results: </strong>The eEEG signals exhibited approximately ×3.7 higher power than concurrent scalp EEG, and nearest endovascular-scalp electrode pairs showed consistently higher coupling across all participants (mean difference 4.9 percentage points, range 1.8-7.6% across individuals), with effects most pronounced at distances $< $30 mm.</p><p><strong>Conclusion: </strong>These findings support the feasibility of eEEG for neuromonitoring and demonstrate its potential for simple brain-computer interface (BCI) applications.</p><p><strong>Significance: </strong>This work provides quantitative measures of the signal power and correlation with scalp EEG, obtained directly in humans for a microcatheter-deliverable wire electrode, establishing human operating bounds for endovascular EEG as a minimally invasive neural interface.</p>","PeriodicalId":13245,"journal":{"name":"IEEE Transactions on Biomedical Engineering","volume":"PP ","pages":""},"PeriodicalIF":4.4,"publicationDate":"2026-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148880212","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":"EMBC Special Issue: PhyNet: A Physiology-Constrained Monotonic Neural Network for Safe and Explainable Blood Glucose Forecasting.","authors":"Andrea Calzavara, Francesco Prendin, Giacomo Cappon, Simone Del Favero, Andrea Facchinetti","doi":"10.1109/TBME.2026.3730324","DOIUrl":"https://doi.org/10.1109/TBME.2026.3730324","url":null,"abstract":"<p><strong>Objective: </strong>Deep learning (DL) has become state-of-the-art for blood glucose (BG) forecasting in type 1 diabetes (T1D). However, its black-box nature raises safety and reliability concerns regarding its use for therapeutic decision-support. This study aims to: (1) highlight potential risks associated with standard DL-based BG forecasting, and (2) address them with PhyNet, a physiology-constrained monotonic neural network.</p><p><strong>Methods: </strong>Two large-scale datasets (T1DEXI and MetaboNet, 848 subjects in total) were used to develop PhyNet-which enforces physiological consistency through a multi-branch structure and weight constraints-and compare it against six DL baselines (convolutional, recurrent, and transformer-based architectures). Models predicted BG levels up to 90-minute ahead using continuous glucose monitoring (CGM) data, carbohydrate intake, and insulin dosing, and were assessed for: (i) predictive accuracy with standard metrics, and (ii) adherence to physiological principles (i.e., carbohydrates increase BG, insulin lowers it) using explainable AI. Specifically, we evaluated model-predicted responses to varying carbohydrate and insulin intakes, and generated counterfactual explanations to identify model-recommended actions for avoiding adverse events.</p><p><strong>Results: </strong>At a 30-minute horizon, predictive accuracy was similar across models (RMSE: 19.39-21.00 mg/dL; Time Gain: 10.45-13.35 min). Despite this, only PhyNet consistently captured the physiological effects of carbohydrates and insulin, yielding 0% unsafe recommendations versus up to 64.3% for baselines.</p><p><strong>Conclusion: </strong>Standard DL models can achieve state-of-the-art performance while failing to respect physiology, posing clinical risk. PhyNet preserves accuracy while enhancing physiological fidelity, supporting safer integration into T1D technologies.</p><p><strong>Significance: </strong>Evaluating physiological consistency alongside predictive accuracy is essential for responsible clinical translation of DL-based BG forecasting.</p>","PeriodicalId":13245,"journal":{"name":"IEEE Transactions on Biomedical Engineering","volume":"PP ","pages":""},"PeriodicalIF":4.4,"publicationDate":"2026-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148880155","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":"Keeping Medical AI Healthy and Trustworthy: A Review of Detection and Correction Methods for System Degradation.","authors":"Hao Guan, David Bates, Li Zhou","doi":"10.1109/TBME.2025.3642706","DOIUrl":"10.1109/TBME.2025.3642706","url":null,"abstract":"<p><p>Artificial intelligence (AI) is increasingly integrated into modern healthcare, offering powerful support for clinical decision-making. However, in real-world settings, AI systems may experience performance degradation over time, due to factors such as shifting data distributions, changes in patient characteristics, evolving clinical protocols, and variations in data quality. These factors can compromise model reliability, posing safety concerns and increasing the likelihood of inaccurate predictions or adverse outcomes. This review presents a forward-looking perspective on monitoring and maintaining the \"health\" of AI systems in healthcare. We highlight the urgent need for continuous performance monitoring, early degradation detection, and effective self-correction mechanisms. The paper begins by reviewing common causes of performance degradation at both data and model levels. We then summarize key techniques for detecting data and model drift, followed by an in-depth look at root cause analysis. Correction strategies are further reviewed, ranging from model retraining to test-time adaptation. Our survey spans both traditional machine learning models and state-of-the-art large language models (LLMs), offering insights into their strengths and limitations. Finally, we discuss ongoing technical challenges and propose future research directions. This work aims to guide the development of reliable, robust medical AI systems capable of sustaining safe, long-term deployment in dynamic clinical settings.</p>","PeriodicalId":13245,"journal":{"name":"IEEE Transactions on Biomedical Engineering","volume":"PP ","pages":"2986-3001"},"PeriodicalIF":4.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13050583/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145722683","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Gas Embolism on a Chip: Mimicking Iatrogenicity at the Microscale.","authors":"K Baassiri, L Maskova, M M Mardanpour, T Allen, F Gamper, L A Flores-Asomoza, A Sudalaiyadum Perumal, D V Nicolau","doi":"10.1109/TBME.2026.3729588","DOIUrl":"https://doi.org/10.1109/TBME.2026.3729588","url":null,"abstract":"<p><strong>Objective: </strong>Gas embolism (GE) is a potentially life-threatening condition with difficult in vivo investigation. The physical nature of the initial stages of GE makes in vitro abiotic microfluidics valuable for phenomenological studies. A versatile microfluidic system was used to mimic the microscale iatrogenic GE during gastroscopy, laparoscopy, and hyperbaric therapy under three scenarios: gas transfer through perforated or non-perforated microvasculature, and via tissue supersaturation.</p><p><strong>Methods: </strong>Polydimethylsiloxane devices mimicking microvascular geometries, with, or without injuries, hosted the flow of artificial blood. Localized gas pressures comparable to those used in gastroscopy, laparoscopy, and hyperbaric therapy were applied on microvasculature using air, carbon dioxide or /nitrogen to mimic microscale GE.</p><p><strong>Results: </strong>Air and carbon dioxide produced markedly different GE patterns. Air generated numerous small bubbles across the entire pressure range, and a distinct population of larger emboli at lower pressures. Conversely, carbon dioxide embolism occurred beyond a threshold pressure (23-58 mm Hg), but once initiated, complete blockage or continuous gas injection lasting tens of seconds occurred.</p><p><strong>Conclusion: </strong>Air embolism may occur in injured microvasculature at pressures within the lower range encountered during gastroscopy. Carbon dioxide embolism requires higher pressures, but once triggered, it can result in high-volume gas entry during laparoscopy. High pressure gas-supersaturated tissues can act as gas 'reservoirs' prolonging GE therapy.</p><p><strong>Significance: </strong>These findings underscore the risk of air GE in gastroscopy and the need for pressure and flowrate control; possible carbon dioxide GE during laparoscopy warrants further assessment; and the duration of gas clearance during GE therapy needs to be considered.</p>","PeriodicalId":13245,"journal":{"name":"IEEE Transactions on Biomedical Engineering","volume":"PP ","pages":""},"PeriodicalIF":4.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148873941","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":"Fiber-Less, Large-Scale Opto-Electrophysiology Interface for Micro-Scale Interaction of Multiple Brain Regions.","authors":"Sungjin Oh, Jose Roberto Lopez Ruiz, Kanghwan Kim, Nathan Slager, Eunah Ko, Mihaly Voroslakos, Hyunsoo Song, Wangbo Chen, Sung-Yun Park, Euisik Yoon","doi":"10.1109/TBME.2025.3646326","DOIUrl":"10.1109/TBME.2025.3646326","url":null,"abstract":"<p><strong>Objective: </strong>Recent neuroscientific research craves for understanding sophisticated brain networks formed by neuron ensembles across multiple regions. An ideal way to unveil the complex connectome is bidirectionally interacting (simultaneous recording and stimulation) with neurons at high spatiotemporal resolutions. Existing CMOS recording probes cannot provide micro-scale interactions with limited stimulation capability. Although optogenetics can achieve neuron-specific stimulation, conventional methods using optic fibers illuminate a large volume of tissue, resulting in unspecific perturbations. While our previous studies demonstrated micro-LED (µLED)-based optoelectrode for localized stimulation and recording, this work advances them into a fully integrated headstage combining the optoelectrode, CMOS IC, and flexible interposer for miniaturized implementation. The proposed system enables micro-scale interactions with high spatiotemporal precision through densely packed 256-neuron-size recording and 128-soma-size fiber-less opto-stimulation across multiple brain regions.</p><p><strong>Methods: </strong>Such high resolutions yet wide coverage is achieved by (1) advanced micromachining techniques integrating recording electrodes and µLEDs, (2) micro-second, independent 384-channel interaction via a low-power, area-efficient circuit, and (3) compact and reliable polyimide-cable-based hybrid assembly.</p><p><strong>Results: </strong>A compact (23.8×28.8 mm²) and lightweight (3.5-gram) headstage achieved the highest reported channel density in area (0.56 channels/mm²) and weight (109.71 channels/gram). A single acute in vivo experiment on a transgenic mouse identified >160 isolated pyramidal neurons and narrow/wide interneurons in the dorsal hippocampus, with local and broad-range effects from focal optogenetic stimulation.</p><p><strong>Conclusion and significance: </strong>We implemented the hybrid integrated, large-scale opto-electrophysiology interface prototype and verified its feasibility in vivo, representing the first fully integrated platform extending our µLED-based probes into a complete system.</p>","PeriodicalId":13245,"journal":{"name":"IEEE Transactions on Biomedical Engineering","volume":"PP ","pages":"3083-3095"},"PeriodicalIF":4.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145793768","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":"Contrastive Decoupling and Enhancement of Multi-view EEG Features for Imagined Speech Decoding.","authors":"Zijian Han, Zhaohu Liu, Honggang Liu, Yong Peng, Li Zhu, Wanzeng Kong, Andrzej Cichocki","doi":"10.1109/TBME.2026.3729643","DOIUrl":"https://doi.org/10.1109/TBME.2026.3729643","url":null,"abstract":"<p><p>Imagined speech decoding remains challenging in brain-computer interfaces (BCIs) due to the low signal-to-noise ratio and complex spatio-temporal-spectral structure of Electroencephalogram (EEG) data. Existing studies mainly rely on single-view features or simple fusion strategies, limiting their ability to capture diverse neural characteristics during speech imagery. To address this limitation, we propose a Multi-view Feature Contrastive Decoupling and Enhancement (MFCDE) framework that integrates multi-view feature construction, feature decoupling, and adaptive masking. Four complementary views, including temporal, frequency-domain, phase-locking value (PLV), and graph-theoretic features, are extracted to characterize speech imagery-related neural dynamics. The decoupling mechanism reduces cross-view redundancy while preserving the discriminative information of each view. Experiments show that MFCDE consistently outperforms existing baselines in classification performance and stability. The learned view-shared and view-specific representations further provide neurophysiological insights by revealing the complementary contributions of temporal, spectral, and connectivity-based EEG patterns to imagined speech discrimination, indicating that reliable decoding depends on the joint utilization of neural dynamics, oscillatory activity, and inter-regional functional interactions.</p>","PeriodicalId":13245,"journal":{"name":"IEEE Transactions on Biomedical Engineering","volume":"PP ","pages":""},"PeriodicalIF":4.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148873947","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":"Predicting Histotripsy Focal Shifts in the Liver From Acoustic Aberrations Using a Deep Learning Model.","authors":"Katrina L Falk, Paul F Laeseke, Ellen Yeats, Timothy L Hall, Grace M Minesinger, Michael A Speidel, Timothy J Ziemlewicz, Fred T Lee, Martin G Wagner","doi":"10.1109/TBME.2026.3729564","DOIUrl":"https://doi.org/10.1109/TBME.2026.3729564","url":null,"abstract":"<p><strong>Objective: </strong>To develop a deep learning model for predicting histotripsy focal shifts in the liver caused by acoustic aberrations for real-time treatment optimization.</p><p><strong>Methods: </strong>A modified VGG19 CNN regression model was trained using 12,870 scenarios derived from 243 segmented human CT volumes. Input to the model consisted of 6-channel maps representing the distance through specific tissue types (bone, air, fat, muscle, liver, total tissue) along transducer-to-focus rays. Ground truth focus locations were determined via acoustic simulations (k-Wave) based on the minimum pressure location. The network was trained to output predicted focal shifts relative to the geometric focus location. Accuracy was evaluated as the mean absolute deviation between CNN predictions and ground truth simulations. An ablation analysis determined dominant features.</p><p><strong>Results: </strong>The simulation predicted aberration-induced focal shifts ranging from -12.9 to 3 mm. Across five-fold cross-validation, the model predicted shifts with a mean absolute deviation (standard deviation) across the folds of 0.3 (0.3), 0.3 (0.3), 0.5 (0.5) mm in X, Y, and Z, respectively. The corresponding mean bias was 0.10 mm, 0.08 mm, and 0.04 mm. A single one-sided t-test confirmed the absolute prediction errors were significantly less than the 1 mm clinical margin (p $mathbf {< }$ 0.001). Ablation analysis revealed fat, liver, and total tissue as dominant predictors. Notably, CNN inference time was 13.8 ms, compared to a median 10.3 (IQR 3.6) hours for acoustic simulations performed on a high-performance cluster.</p><p><strong>Conclusion: </strong>The proposed CNN accurately predicts aberration shifts in focus location comparable to acoustic simulations with millisecond-scale inference speeds, enabling real-time aberration correction.</p>","PeriodicalId":13245,"journal":{"name":"IEEE Transactions on Biomedical Engineering","volume":"PP ","pages":""},"PeriodicalIF":4.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148874017","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}