Annals of Biomedical Engineering最新文献

{"title":"Turning-Point Impact Compromises the Biomechanical Integrity of Cortical Bone During Oscillating Sawing.","authors":"Han Wang, Urara Satake, Toshiyuki Enomoto","doi":"10.1007/s10439-026-04429-7","DOIUrl":"https://doi.org/10.1007/s10439-026-04429-7","url":null,"abstract":"<p><strong>Purpose: </strong>Oscillating bone sawing is widely used in orthopedic surgery, where preserving the mechanical integrity of the prepared bone surface is essential for implant stability. However, the influence of turning-point impacts on the biomechanical integrity of the prepared bone surface remains unclear. This study aimed to quantify the effects of turning-point impacts on surface integrity and the mechanical properties of cortical bone.</p><p><strong>Methods: </strong>Oscillating sawing experiments were performed on bovine cortical bone using conventional and optimized trajectories. Dynamic sawing forces were measured to quantify turning-point impact events. Surface damage was characterized by scanning electron microscopy and digital microscopy. Local and global mechanical properties were evaluated by flat-punch and conventional compression tests, respectively.</p><p><strong>Results: </strong>The conventional trajectory generated pronounced turning-point impact forces, accompanied by extensive surface cracks and breakout defects. In contrast, the optimized trajectory significantly suppressed turning-point impact forces (p < 0.001) and preserved groove wall surface integrity. Consequently, both the local (p < 0.001) and global (p = 0.001) compressive strengths increased, whereas the elastic modulus remained unchanged. These findings indicate that turning-point impacts primarily reduce failure strength by introducing localized microstructural defects while largely preserving bulk elastic stiffness.</p><p><strong>Conclusion: </strong>Turning-point impacts compromise the biomechanical integrity of cortical bone by reducing the load-bearing capacity of implant-relevant bone surfaces. Suppressing these transient impacts helps preserve implant-relevant bone surface integrity and may improve the stability of the bone-implant interface. These findings provide a biomechanical basis for future real-time impact monitoring and adaptive control in surgical robotics to prevent impact-induced damage.</p>","PeriodicalId":7986,"journal":{"name":"Annals of Biomedical Engineering","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149223198","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":"Hemodynamic and Geometrical Parameters Associated with Carotid Artery Stenosis: A Cross-Sectional Study Using 4D Flow MRI and CFD.","authors":"Ali Mokhtari, Ariel Bergmann, Christoph Strecker, Jonathan Andrae, Andreas Harloff, Dominik Obrist","doi":"10.1007/s10439-026-04417-x","DOIUrl":"https://doi.org/10.1007/s10439-026-04417-x","url":null,"abstract":"<p><strong>Purpose: </strong>Carotid geometry and local hemodynamics are closely associated with atherosclerosis, yet how anatomy and disturbed flow jointly mark disease presence and relate to pro-atherogenic flow in still-non-stenosed bifurcations remains unresolved. This study quantifies the cross-sectional association between three-dimensional carotid bifurcation geometry and pro-atherogenic flow and identifies anatomical biomarkers for non-invasive risk stratification.</p><p><strong>Methods: </strong>Geometric analyses included 346 carotid bifurcations, comprising 240 stenosed and 106 non-stenosed bifurcations. Patient-specific CFD was performed in 176 bifurcations: 101 stenosed and 75 non-stenosed. Eight geometric descriptors, including a new offset flow divider capturing bifurcation apex asymmetry, and four ICA bulb hemodynamic indices (TAWSS, OSI, RRT, BFI) were extracted; logistic and linear regression assessed the association of geometry with (i) stenosis presence and (ii) pro-atherogenic flow.</p><p><strong>Results: </strong>Bifurcation angle, ICA angle, tortuosity, and offset flow divider were significantly elevated in stenosed bifurcations (all <math><mrow><mi>p</mi> <mo><</mo> <mn>0.001</mn></mrow> </math> ); a four-feature geometric model showed moderate discrimination between stenosed and non-stenosed bifurcations (cross-validated <math><mrow><mtext>AUROC</mtext> <mo>=</mo> <mn>0.767</mn></mrow> </math> ). The post-stenotic expansion ratio was more strongly associated with local flow disturbance than the degree of luminal narrowing. In non-stenosed bifurcations, multivariable geometric models explained 49 to <math><mrow><mn>60</mn> <mo>%</mo></mrow> </math> of the variance in the flow-normalized low-TAWSS, elevated-OSI, and elevated-RRT outcomes, with bulb area and area ratio providing the largest independent contributions, indicating that geometric variation is strongly associated with candidate pro-atherogenic flow patterns even in the absence of detectable luminal narrowing.</p><p><strong>Conclusions: </strong>Carotid geometry carries two complementary signals: markers of disease presence (bifurcation angle, ICA angle) and candidate anatomical substrates of pro-atherogenic flow in still-non-stenosed bifurcations (bulb area, area ratio). Together with the post-stenotic expansion ratio, these geometric features define a compact, image-derivable set of candidate biomarkers for future early carotid risk stratification.</p>","PeriodicalId":7986,"journal":{"name":"Annals of Biomedical Engineering","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149223177","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":"Respiratory Impedance Prediction in Children and Adolescents: A Meta-Modelling Approach.","authors":"Anuva Chowdhury, Md Mokammel Haque, Mathias Baumert","doi":"10.1007/s10439-026-04323-2","DOIUrl":"https://doi.org/10.1007/s10439-026-04323-2","url":null,"abstract":"<p><p>Respiratory oscillometry provides a simple, noninvasive method for assessing pulmonary function in children and is particularly sensitive to small airway obstruction. But its clinical application remains limited due to the lack of widely accepted reference values. In this study, we developed generalized meta-model equations for respiratory resistance and reactance at 5 Hz in children and adolescents using synthetic data generated from published paediatric oscillometric reference equations. A total of 25 reference equations reported since 2000 were analysed to identify the demographic and methodological factors contributing to variability among existing prediction models. Subsequently, a synthetic dataset comprising 70 healthy children and adolescents aged 2-19 years was generated using physiologically realistic anthropometric combinations derived from established paediatric growth references. Pearson correlation analysis revealed that height, age, and weight were significantly associated with both resistance and reactance, and variance inflation factor analysis indicated strong multicollinearity among these variables. Despite differences among published equations, strong agreement was observed across studies, with correlation coefficients exceeding 0.95 for resistance equations and 0.92 for reactance equations, and most predictions remaining within a ± 30% tolerance range. Stepwise multivariable regression was then used to develop generalized meta-model equations, yielding coefficients of determination of approximately 0.84 for both resistance and reactance at 5Hz. The proposed meta-models were further evaluated across different airway impairment states and bronchodilator responses, demonstrating their ability to preserve physiologically meaningful variations in respiratory resistance and reactance. Collectively, these findings support the proposed meta-models as a standardized framework for paediatric respiratory impedance at 5Hz.</p>","PeriodicalId":7986,"journal":{"name":"Annals of Biomedical Engineering","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-10-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149219741","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":"Abdominal Ultrasound Simulation from Semantic Labels Using Paired Label-to-Physics-Based Image Translation.","authors":"Santiago Vitale, Duilio Deangeli, Ignacio Larrabide, José Ignacio Orlando","doi":"10.1007/s10439-026-04411-3","DOIUrl":"https://doi.org/10.1007/s10439-026-04411-3","url":null,"abstract":"<p><strong>Purpose: </strong>Current abdominal ultrasound (US) simulation methods often require CT-based anatomical references for ray-casting, limiting deformation, and pathology variability. We propose a learning-based pipeline trained to predict physics-based images derived from CT scans from semantic labels, enabling controlled simulations without patient-specific CT volumes at inference time.</p><p><strong>Methods: </strong>We introduce a two-stage pipeline that maps anatomical segmentations to realistic US images through a simplified US image. Stage I synthesizes this image from semantic labels using models trained on CT-based ray-casting outputs. Stage II refines it into a realistic US scan using anatomically guided unpaired translation. Deformations and pathologies are generated by editing anatomical maps.</p><p><strong>Results: </strong>We evaluated Pix2Pix and the Semantic Diffusion Model (SDM) in Stage I, followed by segmentation-guided CycleGAN (SG-CycleGAN) refinement in Stage II. SDM significantly outperformed Pix2Pix in morphological metrics, including MAE <math><mo>↓</mo></math> (19.85 vs. 21.65), SSIM <math><mo>↑</mo></math> (0.28 vs. 0.24), and mIoU <math><mo>↑</mo></math> (0.43 vs. 0.29), whereas Pix2Pix yielded better perceptual point estimates (LPIPS <math><mo>↓</mo></math> : 0.17 vs. 0.19; FID <math><mo>↓</mo></math> : 0.32 vs. 0.37; KID <math><mo>↓</mo></math> : 0.25 vs. 0.48).</p><p><strong>Conclusion: </strong>Training paired generative models with physics-based supervision enables approximation of CT-derived ray-casting outputs at inference time directly from semantic labels. Although the pipeline does not require patient-specific CT volumes at inference time, CT-derived segmentations and ray-casting simulations remain necessary to train Stage I. Once trained, the framework enables controllable healthy and pathological simulations through semantic map modification.</p>","PeriodicalId":7986,"journal":{"name":"Annals of Biomedical Engineering","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149206456","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":"Bone Quality Heterogeneity in Mixed Femoral Metastases Revealed by Stimulated Raman Scattering Spectromicroscopy.","authors":"Azin Mirzajavadkhan, Christian Chum Him Lai, Margarete K Akens, Cari M Whyne, Chenxi Qian","doi":"10.1007/s10439-026-04433-x","DOIUrl":"https://doi.org/10.1007/s10439-026-04433-x","url":null,"abstract":"<p><p>Skeletal metastases disrupt bone remodeling and compromise bone quality; however, existing analytical approaches have not provided spatially resolved characterization of chemical heterogeneity between regions of direct tumor involvement and areas distant to the tumor. Stimulated Raman scattering (SRS) spectromicroscopy is a label-free imaging technique that enables quantitative, spatially resolved assessment of tissue composition. This study characterized the organic and mineral phases of osteoblastic bone present in mixed femoral metastases using SRS microscopy, revealing whether tumor and bone interactions influence regions within the bone distant from the metastasis. Mixed metastases were induced in 5- to 6-week-old athymic male rats with ACE-1 canine prostate cancer cells via intracardiac injection. At 21-day post-inoculation, bilateral femora were harvested. Sagittal sections of the right femora were processed for histological analysis and SRS imaging of the organic phase, and undecalcified sections were used to assess the mineral phase. Regions of interest were selected within newly formed osteoblastic bone, bone distant from the tumor, and in distinct samples containing only healthy bone. Osteoblastic regions showed increases in amide I, amide III, and proline signal intensity, and a reduction in phosphate and phenylalanine signal intensities, reflecting enhanced collagen deposition with diminished mineral content and altered protein composition. Bone distant from the tumor displayed elevated proline and reduced phenylalanine signal intensity, supporting the hypothesis that tumor and bone interactions influence bone beyond areas of direct tumor involvement. These findings establish SRS microscopy as a non-destructive tool for detecting spatially resolved biochemical alterations in metastatic bone.</p>","PeriodicalId":7986,"journal":{"name":"Annals of Biomedical Engineering","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149206468","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":"Automated 3D Analysis of Hip Morphology in a Young Adult Population.","authors":"M A Kamphuis, E H G Oei, J Runhaar, D F Hanff, J J Tolk, R Agricola, S M A Bierma-Zeinstra, S Klein, J Hirvasniemi","doi":"10.1007/s10439-026-04418-w","DOIUrl":"https://doi.org/10.1007/s10439-026-04418-w","url":null,"abstract":"<p><strong>Purpose: </strong>Although many factors contribute to the development of hip OA, structural findings, such as acetabular dysplasia, are recognized as strong contributing factors to early degenerative changes. However, conventional two-dimensional imaging cannot fully capture the joint's three-dimensional anatomy. This study aimed to develop and validate an automated MRI-based method for 3D hip morphology analysis.</p><p><strong>Methods: </strong>MRI data from 2,454 participants in the Generation R Cohort (1,199 males; 1,255 females; mean age 18.4 years; mean BMI 22.7 kg/m²) were analyzed. An nnU-Net deep learning model, trained on 40 manually segmented hips, automatically segmented femoral bone, acetabular bone, and femoral and acetabular cartilage. Model performance was assessed using Dice scores. From the segmentations, five categories of morphological features were computed: basic geometric metrics, cartilage volumes, angular measurements, coverage metrics, and joint space width. Sex differences were evaluated using independent t-tests, and BMI-related differences using one-way ANOVA.</p><p><strong>Results: </strong>The model achieved Dice scores of 0.97 for femoral bone, 0.90 for acetabular bone, 0.76 for femoral cartilage, and 0.77 for acetabular cartilage. Reference ranges were established for all biomarkers. Significant differences were observed between males and females and across BMI groups.</p><p><strong>Conclusion: </strong>Automated 3D hip morphology analysis using MRI is feasible and reliable. The generated reference values enhance understanding of hip development and anatomical variation in young adults and support future large-scale, reproducible assessments. All developed code is made openly available at https://gitlab.com/radiology/msk/genr/hip/hip-segmentation-and-biomarkers.</p>","PeriodicalId":7986,"journal":{"name":"Annals of Biomedical Engineering","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149190751","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":"Pressure-Pressure Loop: A Novel Method for Biophysical Characterization of Blood Vessels.","authors":"Zehava Ovadia-Blechman, Moshe Halak, Dor Ben Ishay, Efrat Rushinek Cohen, Shay Maman, Moran Aviv, Neta Rabin","doi":"10.1007/s10439-026-04422-0","DOIUrl":"https://doi.org/10.1007/s10439-026-04422-0","url":null,"abstract":"<p><strong>Purpose: </strong>To develop and evaluate a pressure-only method for localized functional characterization of synthetic vascular models and vascular grafts by capturing their integrated dynamic response and its interaction with stenosis severity.</p><p><strong>Methods: </strong>An in-vitro closed-loop pulsatile flow system tested seven vascular models and grafts (silicone, latex, ePTFE, Dacron) at 0, 50, and 90% lumen narrowing. Two proximal high-fidelity pressure sensors recorded intraluminal pressure waveforms to construct pressure-pressure loops. Loop behavior was evaluated qualitatively through loop morphology and quantitatively using the slope descriptor (α). Uniaxial tensile testing provided independent Young's modulus measurements.</p><p><strong>Results: </strong>Increasing stenosis altered pressure magnitude while characteristic loop patterns remained recognizable across conditions, with severe stenosis modifying specific features without eliminating broader patterns. Silicone models showed broad loops, latex models displayed complex crossing patterns, and vascular grafts exhibited narrow, near-linear loops. Loop morphology and the repeatable slope descriptor (α) were consistent across measurements, together differentiating the integrated, configuration-specific dynamic responses of the tested vascular models and grafts.</p><p><strong>Conclusion: </strong>Pressure-pressure loop analysis provides a novel pressure-only framework for functional characterization of vascular models and grafts. Combining qualitative loop morphology with a reproducible quantitative slope descriptor, the approach captures complementary features of the integrated dynamic pressure response rather than a single intrinsic material property. The study demonstrates the feasibility of extracting meaningful descriptors from two intraluminal pressure waveforms in-vitro, supporting future validation in native blood vessels and applications in vascular graft design.</p>","PeriodicalId":7986,"journal":{"name":"Annals of Biomedical Engineering","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149178967","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":"Strain-Based Assessment of Post-TAVR Left Bundle Branch Block in Bicuspid Aortic Valve Using Patient-Specific Computational Modeling.","authors":"Salwa Anam, Brandon Kovarovic, Puja Parikh, Ashraf Hamdan, Rami Haj-Ali, Danny Bluestein","doi":"10.1007/s10439-026-04399-w","DOIUrl":"https://doi.org/10.1007/s10439-026-04399-w","url":null,"abstract":"<p><strong>Purpose: </strong>Post-transcatheter aortic valve replacement (TAVR) left bundle branch block (LBBB) is the most common post-TAVR cardiac conduction abnormality (CCA), particularly in bicuspid aortic valve (BAV). Prior computational studies evaluated generalized CCA risk but did not quantify LBBB-specific mechanical injury or compare self-expanding (SE) and balloon-expandable (BE) devices in BAV. This proof-of-concept study established a patient-specific computational framework to (1) quantify mechanical stretch in the left bundle branch (LBB)-adjacent region, (2) identify a preliminary, cohort-specific strain-based separation value associated with post-TAVR LBBB, and (3) compare device-related LBBB risk trends between SE and BE devices.</p><p><strong>Methods: </strong>Cardiac CT scans from six type I BAV TAVR patients were analyzed (two with LBBB and four without CCA). Patient-specific finite element models incorporated anatomy and leaflet calcification. The LBB region was defined using membranous septum length and anatomical landmarks. Simulations reproduced implantation of deployed SE devices and virtual BE devices at standardized depth. Area-weighted logarithmic strain (LE), contact pressure, oversizing index, eccentricity index, and expansion metrics were computed.</p><p><strong>Results: </strong>LBB-adjacent LE showed clear separation (0.15-0.24 in LBBB vs. 0.03-0.11 without CCA), with 0.11 as a preliminary cohort-specific separation value. Mean LE was higher with SE than BE devices (0.11 vs. 0.08). Contact pressure showed no consistent association with outcome, suggesting that LE differences reflect multiple contributing factors.</p><p><strong>Conclusion: </strong>In this six-patient cohort, LBB-adjacent strain clearly separated LBBB from No-LBBB cases, whereas contact pressure showed no consistent relationship with outcome. Larger cohorts and sensitivity analyses are needed to establish its clinical utility for BAV TAVR risk stratification or device selection.</p>","PeriodicalId":7986,"journal":{"name":"Annals of Biomedical Engineering","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149179014","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":"The Efficiency-Hemocompatibility Paradox: A Comparative Lagrangian and Eulerian Analysis of Micro-axial p-LVAD Architectures.","authors":"Yuval Gabso, Moshe Rosenfeld, Idit Avrahami","doi":"10.1007/s10439-026-04406-0","DOIUrl":"https://doi.org/10.1007/s10439-026-04406-0","url":null,"abstract":"<p><strong>Background: </strong>Percutaneous Left Ventricular Assist Devices (p-LVADs) require a delicate balance between hydraulic efficiency and hemocompatibility. While Eulerian turbulence models offer computational speed, their reliability in predicting mechanical blood damage in complex micro-axial geometries compared to established Lagrangian frameworks remains poorly understood.</p><p><strong>Methods: </strong>We employed computational fluid dynamics to evaluate four p-LVAD configurations across 20 operating conditions. Models 1-3 utilized a discharge-type architecture, while Model 4 utilized a suction-type configuration with an expanding hub. Hemocompatibility was assessed using two Eulerian turbulence-based indices (HI₁, HI₂) and two Lagrangian metrics: Median Stress Accumulation (Median SA) and Mean Supra-threshold Exposure Time (t_ATA).</p><p><strong>Results: </strong>A critical design paradox was identified in the suction-type pump: while the expanding hub improved hydraulic efficiency by preventing flow separation, it created a high-velocity bottleneck that significantly elevated bulk shear. Consequently, higher hydraulic efficiency in suction designs was associated with increased Lagrangian blood damage. Eulerian indices functioned primarily as \"RPM trackers,\" showing a near-perfect correlation with the Womersley number (R > 0.94) but failing to capture geometry-induced stress peaks. In contrast, Lagrangian metrics provided a sensitive, multivariate assessment of mechanical loading.</p><p><strong>Conclusion: </strong>Eulerian models are insufficient for evaluating micro-axial pumps where bulk shear is the dominant damage mechanism. Lagrangian particle tracking (median SA and t_ATA) remains the necessary high-fidelity standard for p-LVAD optimization. These findings highlight the fluid dynamic advantages of discharge-type pump architectures, offering a compelling engineering strategy to mitigate high-shear stress exposure compared to suction devices.</p>","PeriodicalId":7986,"journal":{"name":"Annals of Biomedical Engineering","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149178395","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":"Data-Driven Prediction of Speech-in-Noise Performance: A Systematic Review and Meta-analysis of Machine Learning Models.","authors":"Prabhani Athukorala, Anu Nair, Srikanta K Mishra","doi":"10.1007/s10439-026-04407-z","DOIUrl":"https://doi.org/10.1007/s10439-026-04407-z","url":null,"abstract":"<p><strong>Purpose: </strong>Difficulty understanding speech in noise (SIN) is a major determinant of communication ability and quality of life, affecting individuals with hearing loss as well as some with normal hearing. Data-driven prediction methods offer a promising approach for modeling complex, multidimensional factors underlying speech perception. This systematic review and meta-analysis evaluated computational approaches for predicting SIN-related outcomes, compared predictive performance across model classes where quantitative synthesis was feasible, and identified methodological factors relevant to model selection, generalizability, and clinical translation.</p><p><strong>Methods: </strong>Following PRISMA 2020 guidelines, we conducted a systematic review and meta-analysis of studies using machine learning (ML), neural networks, automatic speech recognition (ASR), or auditory model-based approaches to predict, simulate, or evaluate SIN-related outcomes. Quantitative synthesis was performed using random-effects models for studies reporting support vector machine (SVM) or random forest (RF) performance, and between-study heterogeneity was assessed.</p><p><strong>Results: </strong>Eligible studies included traditional ML, neural network, ASR-based, and mechanistic auditory modeling approaches. SVM accuracy pooled at 0.71 and SVM area under the curve at 0.84, while RF accuracy pooled at 0.86 and RF recall at 0.81. Bootstrap estimates were broadly consistent with the primary random-effects estimates, although between-study heterogeneity ranged from moderate to very high across analyses.</p><p><strong>Conclusion: </strong>Data-driven prediction approaches show promise for predicting SIN-related outcomes, but high heterogeneity precludes identifying a single optimal model. Model selection should reflect the prediction target, data type, sample size, interpretability, and intended use. Current evidence supports proof-of-concept performance more strongly than clinical utility, highlighting the need for larger, standardized, externally validated studies.</p>","PeriodicalId":7986,"journal":{"name":"Annals of Biomedical Engineering","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149164421","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}