Physical and Engineering Sciences in Medicine最新文献

{"title":"A novel method of determining a body surface for collision detection using CT localization radiograph.","authors":"Yusuke Ochi, Akito S Koganezawa, Takeo Nakashima, Yuji Murakami","doi":"10.1007/s13246-026-01736-0","DOIUrl":"10.1007/s13246-026-01736-0","url":null,"abstract":"<p><p>In previous studies, identifying a position at high collision risk outside a range of CT imaging requires a cost, such as an infrared camera. In this study, we propose a simple method to identify the patient's body surface using localization radiographs in a planning CT that are available at any facility. Five metallic markers were attached to the mannequin's right elbow with the arm up. Twelve localization radiographs were acquired at 30° intervals from 0° to 330° of the tube angle. The images acquired at 60°, 90°, 120°, 150°, 180°, 270°, 300°, and 330° of the tube angle were used to estimate the metallic markers' position. The three-dimensional (3D) estimated coordinate (EC) of the metallic markers (X, Y, Z) was calculated from the two selected localization radiographs and compared to the reference coordinate (RC). The average of the difference between the EC and the RC was [Formula: see text] -  1.4 ± 3.5 mm, [Formula: see text] = - 3.1 ± 6.2 mm and [Formula: see text] = 0.8 ± 0.5 mm. When the set of combined tube angles (α, β) was (60°,270°), (120°,300°), and (180°,330°), the average distance between RC and EC exceeded 10.0 mm. The maximum distance between RC and EC was 46.7 mm. The distance between the mannequin's elbow and the gantry surface estimated by the TPS differed by 0.2 cm compared with the direct measurement in the treatment room. This study proposed a novel method for estimating patient body surface coordinates using localization radiographs, with sufficient accuracy to prevent collisions between the patient's body surface and the gantry.</p>","PeriodicalId":48490,"journal":{"name":"Physical and Engineering Sciences in Medicine","volume":" ","pages":"1241-1250"},"PeriodicalIF":2.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147934686","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A novel hierarchical transfer learning to enhance breast cancer detection on screening mammography using deep learning.","authors":"Arman Monajemi Mamaghani, Fatemeh Baharvand Ahmadi, Zeinab Naseri, Davood Sotoude, Amin Amiri Tehranizadeh","doi":"10.1007/s13246-026-01781-9","DOIUrl":"10.1007/s13246-026-01781-9","url":null,"abstract":"<p><p>Accurate and early detection of breast cancer in screening mammography remains a critical challenge in medical imaging, particularly due to variations in image quality, tissue density, and annotation standards across datasets. This study aims to design and validate a Hierarchical Transfer Learning (HTL) framework that leverages heterogeneous mammography datasets and multi‑stage pre-processing to achieve more accurate and generalizable deep learning-based detection of breast lesions in screening mammography. Our approach integrates five image enhancement techniques, CLAHE, Gamma Correction, Adaptive Histogram Equalization, Global Histogram Equalization, and Unsharp Masking into the learning process, allowing the model to adapt progressively to enhancement specific features. We evaluate the framework using two state-of-the-art object detection architectures, YOLOv8 and Faster R-CNN, across three publicly available datasets (INbreast, CBIS-DDSM, and MIAS) and a locally collected dataset. Our HTL model achieves a peak mean Average Precision (mAP) of 0.981 and accuracy of 0.998, outperforming previous methods in both generalizability and precision. Moreover, the model maintains robust performance on real-world local data, demonstrating its practical utility for deployment in diverse clinical environments. These results confirm that hierarchical learning combined with strategic image enhancement and pre-processing significantly improves detection performance and supports large scale breast cancer screening applications with robust cross-dataset generalization, enabling consistent performance across heterogeneous imaging domains with varying acquisition protocols, image quality, and annotation standards.</p>","PeriodicalId":48490,"journal":{"name":"Physical and Engineering Sciences in Medicine","volume":" ","pages":"1777-1789"},"PeriodicalIF":2.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148622204","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Knowledge-based treatment planning in breast cancer radiotherapy: comparing different machine learning algorithms.","authors":"Mostafa Robatjazi, Saba Ordibeheshti, Atefeh Rostami, Seyed Alireza Javadinia, Hamid Reza Baghani, Amir Alizadeh","doi":"10.1007/s13246-026-01745-z","DOIUrl":"10.1007/s13246-026-01745-z","url":null,"abstract":"<p><p>Knowledge-based planning (KBP) has emerged as a promising approach to improve the consistency and efficiency of radiotherapy by utilizing prior clinical data to guide treatment planning. This study aimed to evaluate the performance of various machine learning (ML) algorithms for predicting key dosimetric parameters in 3D conformal radiotherapy (3D-CRT) of left-sided breast cancer, with the goal of reducing manual trial-and-error during plan optimization. A retrospective dataset of 75 breast cancer patients treated with 3D-CRT was used. Geometric and dosimetric features, including beam and organ-at-risk (OAR) parameters, were extracted from clinically approved plans. Ten supervised regression algorithms, including SVR, LASSO, Ridge, AdaBoost, Gradient Boosting Regressor, Histogram-Based Gradient Boosting, KNN, MLP, SGD, and Kernel Ridge Regression, were trained to predict dosimetric outcomes such as D_mean and V_x for the heart and lung, and homogeneity index (HI). Data preprocessing included outlier capping, quantile-based oversampling, and MaxAbs scaling. Model performance was evaluated using fivefold cross-validation and an independent test set, employing RMSE, MAPE, and MedAE as metrics. Among the evaluated algorithms, the KNN model demonstrated the consistent predictive performance across all dosimetric endpoints, achieving the lowest RMSE and MAPE for D_mean Heart (2.09 Gy, 0.17) and D_mean Lung (3.12 Gy, 0.15). Feature importance analysis identified geometric parameters such as borders, wedge angles, and OAR volumes as the most influential predictors. ML-based KBP can accurately predict dosimetric outcomes prior to dose calculation, improving planning efficiency and consistency in breast 3D-CRT. The KNN algorithm showed the highest reliability, suggesting its suitability for integration into clinical decision-support systems.</p>","PeriodicalId":48490,"journal":{"name":"Physical and Engineering Sciences in Medicine","volume":" ","pages":"1355-1368"},"PeriodicalIF":2.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148200699","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Distinguishing model- and environment-related effects in knowledge-based planning of moderately hypofractionated prostate VMAT.","authors":"Yuto Yamaji, Motoharu Sasaki, Yuji Nakaguchi, Takeshi Kamomae, Yuki Kanazawa, Kanako Sakuragawa, Ryosuke Kasai, Yuki Tominaga, Hitoshi Ikushima","doi":"10.1007/s13246-026-01778-4","DOIUrl":"10.1007/s13246-026-01778-4","url":null,"abstract":"<p><p>Knowledge-based planning (KBP) can improve the efficiency and consistency of volumetric-modulated arc therapy (VMAT); however, the extent to which treatment-machine characteristics influence KBP model behavior remains unclear. This study aimed to clarify the dosimetric impact of machine-specific KBP models by distinguishing the contributions of model libraries and plan-generation environment. Three RapidPlan models were independently trained using 30 manually generated plans created on NovalisTx (15X), TrueBeam (10X), and Halcyon (6FFF). The models were applied to an independent cohort of 15 patients using two complementary evaluation approaches: a clinical evaluation employing each machine's native beam model and a controlled evaluation using a common TrueBeam 10X beam model with identical optimization settings. Dosimetric endpoints included planning target volume excluding the rectum (PTV-R) D2% and D98%, rectal V31/V46/V54/V58 Gy, and bladder V31/V50 Gy. In the library plans, significant differences were observed for the PTV-R metrics, rectal V54/V58 Gy, and bladder V50 Gy. Under clinical conditions, significant differences remained for the PTV-R D2% and D98% and for bladder V50 Gy, whereas rectal indices showed no significant differences. Under controlled conditions, differences in PTV-R and rectal metrics were no longer significant, whereas bladder V31 Gy and V50 Gy remained significantly different. These findings suggest that the impact of machine-specific KBP models is endpoint-dependent and reflect the combined influence of model-derived objectives and the plan-generation environment. Standardization of beam modeling and optimization conditions reduced inter-model differences for target and rectal metrics, whereas bladder-related differences persisted, indicating that KBP model management in multi-machine environments should consider endpoint-dependent behavior.</p>","PeriodicalId":48490,"journal":{"name":"Physical and Engineering Sciences in Medicine","volume":" ","pages":"1747-1756"},"PeriodicalIF":2.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148551074","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Personalized tricuspid valve modeling predicts regurgitation biomechanical repair and double‑orifice flow.","authors":"Giuseppe Sausa, Giovanni Gentile, Nicola Cuscino, Salvatore Pasta","doi":"10.1007/s13246-026-01740-4","DOIUrl":"10.1007/s13246-026-01740-4","url":null,"abstract":"<p><p>Tricuspid regurgitation (TR) is a prevalent and progressive condition associated with heart failure symptoms; however, numerous patients are unsuitable for surgery, and achieving a long-term outcome is challenging. This renders TR an unmet clinical problem. Computational modeling facilitates the evaluation of the biomechanics associated with regurgitation, aiming to enhance therapeutic strategies as transcatheter edge-to-edge repair. This study sought to establish a computational framework integrating a tricuspid valve (TV) model with the right ventricle (RV) wall to evaluate native valve function, TR pathology, and repair using the MitraClip device. Computed tomography angiography (CTA) from three patients was utilized to reconstruct the tricuspid valve leaflets, annulus, and RV wall. A parametric CAD pipeline for modeling the leaflet surfaces and chordal architecture was built using ex-vivo data. Annular kinematics (systole-to-diastole) were computed and then applied as a boundary condition in the finite-element analyses, with tricuspid regurgitation induced by displacing papillary positions and augmenting transvalvular loading. The MitraClip deployment was subsequently simulated whereas the post-deformation hemodynamics were assessed using a lattice-Boltzmann solver to quantify intraventricular velocity fields and vorticity. All three models demonstrated physiological leaflet coaptation and stress distributions in a healthy patient condition, with maximum stresses at end-systole quantified in the leaflet belly and chordal insertions. Post-MitraClip simulations demonstrated restored valve closure without retrograde flow during systole; however, flow exhibited a double-orifice inflow and localized vorticity near the clip during diastole. The jet orientation and downstream patterns depended on patient anatomy. This computational framework replicates TV biomechanics and post-repair hemodynamics from standard imaging, facilitating quantitative, patient-specific evaluation. The current patient-specific, image-based computational method can predict TV-related performance and post-repair hemodynamic outcomes, facilitating personalized planning to minimize residual TR and enhance device placement.</p>","PeriodicalId":48490,"journal":{"name":"Physical and Engineering Sciences in Medicine","volume":" ","pages":"1285-1295"},"PeriodicalIF":2.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13638743/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147870095","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Compact neural network algorithm for electrocardiogram classification.","authors":"Christian M Frausto-Avila, Juan P Manriquez-Amavizca, Ana K S Rocha-Robledo, Mario A Quiroz-Juarez, Alfred B U'Ren","doi":"10.1007/s13246-026-01751-1","DOIUrl":"10.1007/s13246-026-01751-1","url":null,"abstract":"<p><p>In this paper, we present a powerful, compact electrocardiogram (ECG) classification algorithm for cardiac arrhythmia diagnosis that addresses the current reliance on deep learning and convolutional neural networks (CNNs) in ECG analysis. This work aims to reduce the demand for deep learning, which often requires extensive computational resources and large labeled datasets. Our approach introduces an artificial neural network (ANN) with a simple architecture combined with a compact, interpretable feature-engineering pipeline. A key contribution of this work is the incorporation of 17 engineered features that enable the extraction of critical patterns from raw ECG signals. By integrating mathematical transformations, signal processing methods, and data extraction algorithms, our model captures the morphological and physiological characteristics of ECG signals with high efficiency, without requiring deep learning. Our method demonstrates a similar performance to other state-of-the-art models in classifying five rhythm classes-normal sinus rhythm, sinus bradycardia, sinus tachycardia, ventricular flutter, and atrial fibrillation. Our algorithm achieved an accuracy of [Formula: see text] on the MIT-BIH and St. Petersburg INCART arrhythmia databases, with a Cohen's kappa of [Formula: see text] and a Matthews correlation coefficient of [Formula: see text]. The compactness of the model and its low inference latency on a standard CPU suggest that the approach is a promising candidate for deployment on resource-constrained platforms.</p>","PeriodicalId":48490,"journal":{"name":"Physical and Engineering Sciences in Medicine","volume":" ","pages":"1393-1404"},"PeriodicalIF":2.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13638727/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148220144","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Expanded methods for geometric location of 2D bone histology co-registration in 3D radiology volumes.","authors":"Jaime Uy De Baron, Robert Phillips, Anthony Butler","doi":"10.1007/s13246-026-01764-w","DOIUrl":"10.1007/s13246-026-01764-w","url":null,"abstract":"<p><p>Bone sarcoma treatment employs wide surgical margins to ensure complete tumour excision, sometimes at the expense of patient quality of life. As precision technologies advance, quantitative data to inform the planned accuracy of surgical margins is increasingly important. This study assesses two novel computational methods (M2, M3) to support the co-registration of 2D histologic images within 3D radiological volumes. Replicated evaluation of an established method (M1) is also performed with new data. M2 suits concave bisected sections by assuming a different orientation during tissue bandsaw than M1. M3 handles freeform geometries by using a third measurement to constrain 2D histology images within 3D space without need for geometric bandsaw assumptions. Three computed distances across 38 canine specimen dissections from each method were compared with three corresponding physical measurements collected during laboratory processing. M1 produced a mean error of 0.12 (± 1.7) mm, M2 produced a mean error of 0.06 (± 1.9) mm, while M3 produced a mean error of 0.02 (± 1.5) mm. Mean errors for each method were of similar magnitude to 3D registration error reported in histology co-registration frameworks for other tissues. This study supports a simple low cost opportunity for use of digital histology images to support bone sarcoma radiology interpretation and treatment planning.</p>","PeriodicalId":48490,"journal":{"name":"Physical and Engineering Sciences in Medicine","volume":" ","pages":"1551-1559"},"PeriodicalIF":2.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13638728/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148594190","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The influence of stenosis severity and geometric orientation on atherosclerotic progression: A fluid-structure interaction study in the carotid artery.","authors":"Harshawardhan Patil, Alexandrina Untaroiu","doi":"10.1007/s13246-026-01784-6","DOIUrl":"10.1007/s13246-026-01784-6","url":null,"abstract":"<p><p>Atherosclerosis remains the leading global cause of mortality, and its progression is strongly influenced by the local hemodynamic environment. Low Wall Shear Stress (WSS) is widely recognized as a promoter of atherosclerosis, yet the combined influence of stenosis severity and its geometric orientation on disease progression in anatomically realistic geometries has not been systematically examined. This study assesses the risk of atherosclerotic progression in a carotid artery using one-way Fluid-Structure Interaction (FSI). Four stenosis levels were simulated (0, 30, 60, and 80%), along with three angular orientations of the stenotic apex relative to the ICA wall. Blood flow was modeled using the [Formula: see text] SST turbulence model, and arterial walls were represented using a Mooney-Rivlin hyperelastic formulation. For 0% and 30% stenosed cases, regions of low Time-Averaged WSS (TAWSS <= 0.4 Pa) appeared in the ICA. In contrast, the 60% and 80% stenoses produced markedly elevated TAWSS at the stenosis throat, ranging from 20-60 Pa. Arterial wall deformation of the 0% and 30% cases was observed around 0.8 mm, while the 60% and 80% stenoses exhibited prominent foci of high deformation upto 2.1 mm. The presence of disturbed flow dynamics and their resulting atheroprogression risk was assessed through the interplay of Reynolds number, nondimensional WSS, nondimensional Turbulent Kinetic Energy (TKE) and the Oscillatory Shear Index (OSI). At 30% severity, the upstream-tilted stenosis intensified early-stage progression indicators, whereas at 60% severity, the downstream-tilted orientation increased plaque vulnerability. The study demonstrates a comparative framework to rank relative risk of atheroprogression using the disturbed flow signatures and wall mechanics observed across the 8 geometries.</p>","PeriodicalId":48490,"journal":{"name":"Physical and Engineering Sciences in Medicine","volume":" ","pages":"1851-1866"},"PeriodicalIF":2.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13638714/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148971754","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A hybrid model for early diagnosis of ophthalmology diseases leveraging CNNs, SBOA optimization, and XAI for visualization.","authors":"Muhammed Yildirim, Mucahit Karaduman, Soner Kiziloluk, Dilber Cetintas, Aziz Aksoy","doi":"10.1007/s13246-026-01771-x","DOIUrl":"10.1007/s13246-026-01771-x","url":null,"abstract":"<p><p>Ophthalmology diseases are among the leading causes of vision loss worldwide. Glaucoma, diabetic retinopathy, and cataracts are the most common diseases and can lead to permanent vision loss if left untreated. In this paper, a new hybrid model has been proposed with the methods accepted in the literature used in the early diagnosis of these diseases. The relationships between imaging analyses and clinical evaluations performed in the diagnostic processes of glaucoma, diabetic retinopathy, and cataract are discussed, and the methods that help to identify diseases in the early stages are emphasized. In addition, the contributions of advanced technologies and imaging systems used in diagnosing these diseases to the developments in the field of eye health are discussed. This article proposes a hybrid model for eye disease detection that combines Convolutional Neural Networks (CNNs) with a metaheuristic optimization algorithm. This model uses ShuffleNet and ResNet101 models as feature extractors, while the Secretary Bird Optimization Algorithm (SBOA) is used for feature selection. Then, the extracted feature maps were combined with ShuffleNet and ResNet101 and optimized with SBOA. The feature fusion process aimed to improve the performance of the developed model by combining different features of the same image. The combined feature map optimized with SBOA was classified into six different classifiers so that the model could work faster and more effectively. Competitive results were produced in the developed model. Finally, explainable artificial intelligence methods were used to visualize the decisions of the developed hybrid model and understand the internal working principle of the model.</p>","PeriodicalId":48490,"journal":{"name":"Physical and Engineering Sciences in Medicine","volume":" ","pages":"1621-1635"},"PeriodicalIF":2.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148473632","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An artificial intelligence and numerical simulation-based strategy for multi-objective optimization of hemodynamics in curved vessels.","authors":"Hamed Khajavi, Amir Rastgoo, Fariborz Masoumi, Farhad Sadegh Moghanlou","doi":"10.1007/s13246-026-01783-7","DOIUrl":"10.1007/s13246-026-01783-7","url":null,"abstract":"<p><p>Understanding and optimizing blood flow behavior in curved vessels is crucial for improving cardiovascular treatment planning and reducing complications such as thrombosis and impaired oxygen transport. Although recent advances in computational modeling have enhanced the simulation of blood rheology, systematic optimization of flow parameters under controlled geometric conditions remains limited. This study addresses this gap by optimizing the hemodynamic performance of curved blood vessels using numerical modeling and multi-objective optimization. Three key parameters hematocrit level, kinematic viscosity, and curvature degree were analyzed to enhance flow velocity and minimize shear stress, balancing oxygen delivery and thrombus formation risks. A numerical model based on the HemoCell framework and the Lattice Boltzmann-Immersed Boundary Method simulated 250 scenarios across ten vessel geometries with varying curvature. Predictive modeling was performed using gradient boosting and support vector regression integrated with bio-inspired optimization algorithms. The Sparrow Search Algorithm combined with gradient boosting achieved the highest velocity prediction accuracy (R² = 0.99902), while the Hunger Game Search Algorithm demonstrated superior performance in shear stress prediction (R² = 0.99577). These results confirmed the reliability of the surrogate models for optimization. Pareto fronts and optimal points were then identified, revealing a shear stress of 587.70 Pa and a velocity of 0.039 m/s as one of the optimal trade-offs. The proposed approach provides a validated framework for hemodynamic optimization and provides preliminary insights that could inform future studies on thrombus risk assessment and the bio-inspired design of bypass grafts.</p>","PeriodicalId":48490,"journal":{"name":"Physical and Engineering Sciences in Medicine","volume":" ","pages":"1817-1849"},"PeriodicalIF":2.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148841203","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}