Annals of Biomedical Engineering最新文献

{"title":"Agentic AI in Medicine: Challenges for Responsible Development and the Case for Clinical Testing Harnesses.","authors":"Ethan Waisberg, Joseph W Guarnieri","doi":"10.1007/s10439-026-04381-6","DOIUrl":"https://doi.org/10.1007/s10439-026-04381-6","url":null,"abstract":"<p><p>Clinical medicine evaluates decision-support technology using methods built around a single output presented for clinician review. Agentic artificial intelligence does not fit this design. Agent scaffolding, the software infrastructure surrounding a large language model that turns passive text generation into active, multi-step execution, allows the model to retrieve information, invoke external tools, and act on the results before a clinician sees any of it. The relevant unit of clinical risk therefore shifts from a single inference to a trajectory of actions. We describe four challenges this creates for responsible development: silent error propagation across multi-step tasks, oversight that reviews conclusions rather than processes, validation that does not survive changes to model or tooling, and scope that widens faster than the evidence supporting it. We argue that each is addressed by a clinical testing harness: a structured evaluation environment comprising scenario libraries built from clinical edge cases, full-trajectory observability, explicit escalation testing, and staged evidence thresholds tied to scope of practice. Medicine already possesses these tools in the form of simulation, credentialling, and morbidity and mortality review, and requires their adaptation rather than their invention.</p>","PeriodicalId":7986,"journal":{"name":"Annals of Biomedical Engineering","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149029598","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":"Ultrasonic Versus High-Speed Drilling for C2 Pedicle Screw Placement with High-Riding Vertebral Artery: A Comparative Study of Vascular Safety and Thermal Injury.","authors":"Changjiang Ou, Yongjun Tong, Qibin Zhang, Bao Huang, Xiangxi Kong, Jiayan Jin, Zimin Cai, Jian Chen, Yilei Chen, Junhui Liu, Fengdong Zhao","doi":"10.1007/s10439-026-04389-y","DOIUrl":"https://doi.org/10.1007/s10439-026-04389-y","url":null,"abstract":"<p><strong>Background and objective: </strong>High-riding vertebral artery (HRVA) anatomy markedly narrows the corridor for C2 pedicle screw placement and increases the risk of vertebral artery injury. Ultrasonic bone drilling (UBD) may improve vascular safety compared with high-speed drilling (HSD), but its thermal effects and influence on early fixation remain unclear. This study compared UBD and HSD for C2 pedicle screw tract preparation in HRVA anatomy.</p><p><strong>Methods: </strong>This integrated clinical-experimental study included a retrospective clinical cohort of 33 patients with 43 HRVA pedicles, a perfused 3D-printed C2 benchtop model with porcine arteries, and a paired rabbit model. Clinical outcomes included conversion to alternative fixation due to failure, substantial venous plexus bleeding, and vascular injury. In the benchtop model, standardized 1-, 2-, and 3-mm pedicle breaches were created to assess arterial injury risk. In rabbits, intraoperative bone temperature, acute osteonecrosis depth, bone healing by micro-CT, and screw pull-out strength at 0, 4, and 12 weeks were evaluated.</p><p><strong>Results: </strong>In the clinical cohort, UBD was associated with lower rates of conversion to alternative fixation (3.8% vs. 35.3%, p = 0.022) and substantial venous bleeding (7.7% vs. 35.3%, p = 0.034); one minor vertebral artery injury occurred in the HSD group, although the low event rate limits statistical power (p = 0.395). In benchtop testing, HSD caused more frequent full-thickness arterial perforation at 2 mm and 3 mm breaches. In rabbits, UBD generated higher peak bone temperatures and deeper acute osteonecrosis, with a trend toward delayed bone healing. Pull-out strength was significantly lower in the UBD group at 4 weeks; at 12 weeks, no statistically significant difference was detected between groups.</p><p><strong>Conclusion: </strong>UBD was associated with improved intraoperative feasibility in this retrospective HRVA cohort and showed fewer vessel perforations in a benchtop model, but prospective validation is required. Its greater thermal injury and lower pull-out strength at 4 weeks highlight the need for optimized drilling parameters and effective irrigation.</p>","PeriodicalId":7986,"journal":{"name":"Annals of Biomedical Engineering","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149029608","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":"Impact of Microvascular Resistance on Coronary Hemodynamic Diagnostic Indices: A Multiscale Simulation Study.","authors":"Tej Jolly, Arnav Garcha, Noelia Grande Gutiérrez","doi":"10.1007/s10439-026-04378-1","DOIUrl":"https://doi.org/10.1007/s10439-026-04378-1","url":null,"abstract":"<p><strong>Purpose: </strong>Coronary diagnostic indices-fractional flow reserve (FFR) and coronary flow reserve (CFR)-disagree in up to 30% of cases, complicating diagnosis and revascularization decisions. Microvascular health likely drives this discordance, but its isolated effect is challenging to quantify in vivo and is often confounded by anatomical factors in computational studies.</p><p><strong>Methods: </strong>Here, we integrated three-dimensional computational fluid dynamics (CFD) with a lumped-parameter network of the coronary microvasculature, systematically varying hyperemic microvascular resistance (HMR) across physiological ranges while keeping epicardial lesion geometry fixed. We simulated nine left coronary artery models with varying lesion locations (LAD/LCx), lengths (1.2/2.5 cm), and stenosis severities (0/45/60%). Each model was evaluated under four hyperemic microvascular resistance states, yielding 36 hyperemic simulations for the FFR, HMR, and HSR analyses. An additional resting simulation was performed for each coronary artery model, yielding nine resting simulations, and 45 simulations in total. Pressure and velocity fields were used to calculate FFR, CFR, HMR, and hyperemic stenosis resistance (HSR), enabling quantification of global relationships and local sensitivities near clinical thresholds (FFR=0.8; CFR=2).</p><p><strong>Results: </strong>Increasing HMR elevated FFR and reduced CFR for identical lesions, suggesting that coronary flow is primarily governed by HMR rather than stenosis severity. Higher HMR modestly increased the arterial wall area exposed to low wall shear stress at bifurcations. Near the clinical thresholds, the local associations of FFR with HMR and HSR were comparable in magnitude, whereas CFR was more strongly associated with HMR than HSR. Cases with low FFR (<0.50) remained hemodynamically significant irrespective of HMR. Both modes of FFR-CFR discordance were reproduced by varying HMR in otherwise identical geometries, demonstrating that HMR alone can significantly change lesion hemodynamic significance.</p><p><strong>Conclusion: </strong>Our findings support reporting HMR alongside FFR and CFR to enhance clinical interpretation and improve physiological accuracy in CFD-based coronary assessment.</p>","PeriodicalId":7986,"journal":{"name":"Annals of Biomedical Engineering","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149007721","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":"Patient-Specific Aortic Growth Prediction in Type B Aortic Dissection Using Finite-Element Simulation Based on the Unified Fiber Distribution (UFD) Model.","authors":"Xue Liang, Marc-Philipp H Schmid, Minliang Liu, Hannah L Cebull, Michael Zhang, Sunny Xu, Muhammad Naeem, John N Oshinski, John A Elefteriades, Rudolph L Gleason, Bradley G Leshnower, Hai Dong","doi":"10.1007/s10439-026-04379-0","DOIUrl":"10.1007/s10439-026-04379-0","url":null,"abstract":"<p><strong>Purpose: </strong>Type B aortic dissection (TBAD) managed with optimal medical therapy (OMT) carries a 40-50% risk of progressive aneurysmal degeneration. Predicting which patients will experience significant aortic enlargement is critical for timely intervention. This study aimed to develop and evaluate a patient-specific finite-element (FE) framework for predicting aortic geometry and diameter growth in uncomplicated TBAD managed with OMT.</p><p><strong>Methods: </strong>We extended our previously developed wall stress-driven growth framework and applied it to seven TBAD patients with each having three serial computed tomography (CT) scans. The unified fiber distribution (UFD) model was applied to describe aortic wall mechanics, and a novel centerline-based algorithm was developed to determine the local material coordinates of aortic tissues. Patient-specific aortic geometries from the three serial CT scans were obtained for each of the seven TBAD patients. Using the first two CT images and each patient's blood pressure, inverse FE analysis was performed to obtain patient-specific growth parameters. These parameters were then used to simulate forward growth and predict geometry at the third time point.</p><p><strong>Results: </strong>Predicted aortic geometries and dimensions matched well with in vivo measurements: for all seven patients, the absolute percent error was < 3.5% for the predicted maximum aortic diameter and < 4% for the predicted mean diameter across all locations.</p><p><strong>Conclusion: </strong>This proof-of-concept study demonstrates the feasibility of patient-specific TBAD growth forecasting from routine CT imaging and blood pressure, which could support individualized surveillance planning and earlier, evidence-based decision-making for patients managed with OMT.</p>","PeriodicalId":7986,"journal":{"name":"Annals of Biomedical Engineering","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149001082","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":"In Silico Modeling and Validation of Post-Dilation in TAVI Patients.","authors":"Letizia Maria Perri, Benedetta Grossi, Virginia Fregona, Sara Barati, Lucia Barbieri, Gabriele Dario Tumminello, Stefano Carugo, Elena Bianchi, Paola De Stefano, Francesca Berti, Ottavia Cozzi, Gianluigi Condorelli, Giulio Stefanini, Gabriele Dubini, José Felix Rodriguez-Matas, Francesco Migliavacca, Giulia Luraghi","doi":"10.1007/s10439-026-04325-0","DOIUrl":"https://doi.org/10.1007/s10439-026-04325-0","url":null,"abstract":"<p><strong>Purpose (stating the main purposes and research question): </strong>The purpose of this study is to develop and validate a balloon post-dilation model into patient-specific transcatheter aortic valve implantation (TAVI) simulations, addressing the current lack of numerical models capable of reproducing this procedural step.</p><p><strong>Methods: </strong>A finite element model of a True^TM Dilatation (BD, Tempe, Arizona) balloon was reconstructed and discretized. Balloon material properties were calibrated and validated through experimental tensile and inflation tests, respectively. The balloon model was then integrated into the final configurations of four TAVI simulations to reproduce the post-dilation phase. Model predictions were quantitatively compared with post-operative computed tomography (CT) images to assess the reliability of the final device configuration. Additionally, simulations performed with and without post-dilation phase were compared to evaluate differences in prediction error and to quantify the impact of post-dilation on agreement with clinical imaging.</p><p><strong>Results: </strong>The calibrated balloon model accurately reproduced experimental inflation behavior around the nominal working pressure, and post-dilation was simulated in all cases. Quantitative comparison with post-operative CTs demonstrated high predictive accuracy of the final stent configuration. The average percentage error in orifice area was 1.55 ± 1.26%, while eccentricity error at the left ventricular outflow tract level was 1.54 ± 1.08. Inclusion of post-dilation reduced prediction errors by 53% for orifice area and 80% for eccentricity.</p><p><strong>Conclusion: </strong>The integration of balloon post-dilation significantly improves the accuracy and clinical relevance of patient-specific TAVI simulations. Incorporating this step is essential to reliably predict final stent configuration and support the development of robust in silico tools for personalized TAVI planning and risk assessment.</p>","PeriodicalId":7986,"journal":{"name":"Annals of Biomedical Engineering","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148975173","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":"Prompt Injection in Clinical Artificial Intelligence Systems: The Emerging Security Challenge of Large Language Models and Agentic AI","authors":"Ethan Waisberg,&nbsp;Joseph W. Guarnieri","doi":"10.1007/s10439-026-04376-3","DOIUrl":"10.1007/s10439-026-04376-3","url":null,"abstract":"<div><p>Clinical discussion of artificial intelligence safety has concentrated on accuracy, bias, and hallucination, each of which describes a model failing at its assigned task. Prompt injection describes the opposite condition: a model performing exactly as instructed, by an instruction the clinician neither wrote nor can see. The vulnerability arises from a fundamental property of current language-model architectures, which receive an undifferentiated stream of tokens and possess no mechanism for distinguishing content that carries authority from content that does not. Medicine is particularly exposed, because the clinical record is assembled from material originating outside the institution, including referral correspondence, patient-entered messages, external reports, scanned documents, and imaging acquired elsewhere. We argue that prompt injection warrants classification as a patient safety hazard with an articulated threat model rather than as a technical curiosity, that improved prompting and input filtering do not address it, and that the workable response lies in provenance-aware context handling, restricted privileges for irreversible actions, and adversarial testing before deployment.</p></div>","PeriodicalId":7986,"journal":{"name":"Annals of Biomedical Engineering","volume":"54 10","pages":"3269 - 3272"},"PeriodicalIF":5.2,"publicationDate":"2026-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148975190","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":"Transfer Learning on Segment Anything Model for Footwear Outsole Segmentation to Predict Footwear Slip Resistance.","authors":"Shaghayegh Chavoshian, Ali Barzegar Khanghah, Atena Roshan Fekr","doi":"10.1007/s10439-026-04331-2","DOIUrl":"https://doi.org/10.1007/s10439-026-04331-2","url":null,"abstract":"<p><strong>Purpose: </strong>Efficient and accurate footwear outsole segmentation is essential in safety-related applications, particularly for understanding outsole-ground interaction to prevent injuries. Identifying contact regions supports applications such as slip resistance evaluation, ergonomic design, and safety standard development. However, analyzing large datasets is time-consuming and costly due to the need for precise manual annotation. The Segment Anything Model (SAM) offers versatile segmentation capabilities, but the complex patterns and textures of outsoles limit its performance, requiring task-specific fine-tuning. This study proposes a fine-tuned SAM model to improve outsole segmentation.</p><p><strong>Methods: </strong>A dataset of 40 footwear outsoles was manually annotated using a graphical tool. The developed segmentation model was used to estimate outsole-ground contact areas, which were then input into a machine learning model to predict slip resistance categories. Ground truth labels were derived from human-centered data, classifying footwear as low or high slip resistance.</p><p><strong>Results: </strong>The fine-tuned SAM model outperformed the original, reducing loss by 8.11%, increasing IoU to 70.45%, improving accuracy from 56.90 to 78.10%, and raising the F1 score from 53.70 to 66.30%. Image quality factors such as resolution, contrast, and intensity significantly affected performance. The classification model achieved 70% accuracy and a 64% F1 score using an 80/20 train-test split.</p><p><strong>Conclusion: </strong>These findings support the use of fine-tuned segmentation models for improving slip resistance assessment, with important implications for preventing falls among older adults and reducing workplace injuries.</p>","PeriodicalId":7986,"journal":{"name":"Annals of Biomedical Engineering","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148944085","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":"Flow-Induced High-frequency Vascular Wall Vibrations: A New Mechanobiological Stimulus for Vascular Remodeling?","authors":"Andrea Remuzzi, Luca Soliveri, Sofia Poloni, Elena Carrara, Chiara Emma Campiglio, Kristian Valen-Sendstad, Michela Bozzetto","doi":"10.1007/s10439-026-04360-x","DOIUrl":"https://doi.org/10.1007/s10439-026-04360-x","url":null,"abstract":"<p><p>Despite substantial progress in vascular access research, the mechanobiological mechanisms driving adverse remodeling of vascular walls, such as intimal hyperplasia, remain unclear. Native arteriovenous fistula (AVF), the preferred vascular access for hemodialysis, represents an ideal condition to investigate the effect of hemodynamic changes on vascular cells' biological functions. Our recent work provides converging evidence that disturbed venous flow in AVFs not only triggers endothelial dysfunction but also induces high‑frequency mechanical vibrations within the vessel wall. These vibrations may directly influence smooth muscle cell proliferation, phenotype, and overall wall remodeling. This review summarizes relevant work in the field and calls for a paradigm shift in future AVF research: moving beyond the role of hemodynamic shear stress on the endothelium and considering the role of vibration‑related mechanobiological stimuli in these pathological processes acting throughout the entire vascular wall and involving all vascular cell phenotypes.</p>","PeriodicalId":7986,"journal":{"name":"Annals of Biomedical Engineering","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148944107","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":"Assessing the Credibility of Quantitative Information: A General Framework","authors":"Marco Viceconti","doi":"10.1007/s10439-026-04367-4","DOIUrl":"10.1007/s10439-026-04367-4","url":null,"abstract":"<div><p>Quantitative information can be obtained through measurement, inference, or prediction. Each of these approaches has its own theoretical framework for assessing the credibility of the information produced: metrology for measured information, statistics for inferred information, and Computational Science and Engineering for predicted information, particularly through the Verification, Validation, and Uncertainty Quantification framework. These credibility assessment frameworks are mature and well-established, and they are especially important in fields such as <i>in silico</i> medicine, where reliable information is essential to support clinical decision-making. However, intensive research in this area is generating new classes of information estimators that do not fit neatly into any of these three categories. Examples include <i>in silico</i>-augmented clinical trials, physics-informed machine learning predictors, and the use of synthetic datasets to train machine learning models. In this letter, we propose a generalisation of these three distinct approaches to the problem, called the 7S Framework, which addresses this gap by integrating and generalising the credibility assessment approaches used in metrology, statistics, and Computational Science and Engineering. We applied the 7S Framework to seven diverse use cases; the results show that the framework is effective, sufficiently general, and capable of capturing the subtle differences that shape the concept of credibility across different types of quantitative information.</p></div>","PeriodicalId":7986,"journal":{"name":"Annals of Biomedical Engineering","volume":"54 10","pages":"3261 - 3268"},"PeriodicalIF":5.2,"publicationDate":"2026-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10439-026-04367-4.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148928828","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":"Exploratory Non-invasive Estimation of Cervical Dilation from Electrohysterography and Maternal Data: A Machine-Learning Proof-of-Concept Study.","authors":"Otniel Portillo-Rodríguez, Jorge Escalante-Gaytán, Oscar O Sandoval-González, Paula Romina Soria, Hugo Mendieta-Zerón, Juan Carlos Echeverría, Miguel Ángel Peña-Castillo, Eric Alonso Abarca-Castro, José Javier Reyes-Lagos","doi":"10.1007/s10439-026-04349-6","DOIUrl":"https://doi.org/10.1007/s10439-026-04349-6","url":null,"abstract":"<p><p>Cervical dilation during labor is commonly assessed through repeated vaginal examinations, which can be uncomfortable, depend on the observer, and are not suitable for continuous monitoring. This pilot study explored whether machine-learning regression could retrospectively estimate cervical dilation using non-invasive abdominal electrohysterography (EHG) recordings and maternal information. Seventy-two 10-min EHG recordings were collected during labor at two Mexican hospitals. Twenty-one EHG features were calculated from nine overlapping 120-s windows and summarized using eight temporal descriptors, generating 168 temporal EHG features. The recording-level low-contraction (LC) and high-contraction (HC) counts, along with maternal and gestational age, were subsequently appended, yielding a 172-feature vector per recording. After removing five recordings with sparse data outside the 4-8 cm dilation range, 67 records remained-50 for model development and 17 for internal testing. In the development phase, feature ranking, screening of 28 regression models, and genetic-algorithm optimization were performed. The genetic algorithm selected five EHG-derived temporal features for a squared-exponential Gaussian process regression model. During fivefold cross-validation within the development set, the optimized model achieved an RMSE of 0.97 cm, r = 0.67, and R² = 0.45. On the held-out internal test set, the RMSE was 1.45 cm, with r = 0.23 and R² =  - 0.31. The model showed moderate cross-validated performance during development but limited generalization in the held-out internal test set. These findings are exploratory and do not establish performance for dystocia screening. Larger prospective studies are needed before clinical use can be considered.</p>","PeriodicalId":7986,"journal":{"name":"Annals of Biomedical Engineering","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148918136","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}