Medical Engineering & Physics最新文献_第9页

An adaptive bin-stream network based on frequency decomposition for classifying atrial fibrillation with low SNR data 基于频率分解的自适应双流网络在低信噪比房颤分类中的应用

IF 2.3 4区医学 Q3 ENGINEERING, BIOMEDICAL

Medical Engineering & Physics

Pub Date : 2025-08-07 DOI: 10.1016/j.medengphy.2025.104412

Jilin Wang , Tengqun Shen , Mengfan Li , Yijun Ma , Guozhen Sun , Yatao Zhang

To detect atrial fibrillation (AF) in ECG signals with low signal-to-noise ratio (SNR), this study introduces the adaptive bin-stream network (ABNet) based on frequency decomposition. The ABNet offers notable advantages: it exhibits high robustness in identifying AF amidst noisy environments, it decomposes the ECG signals into 32-frequency channel recordings to refine frequency ranges for better identifying AF, and it designs an adaptive bin-stream network to gain the optimal results. The method utilizes a 5-level Haar wavelet packet decomposition to decompose the preprocessed ECG signals into their corresponding 32-frequency channel recordings, and the preprocessing signals and the recordings are fed into waveform stream and frequency stream of the bin-stream network, respectively. Finally, an adaptive approach is employed to obtain the optimal classification results. The ABNet was validated for the PhysioNet/Computing in Cardiology Challenge 2017 database (CinC 2017 Db) to classify 4 categories i.e., normal sinus rhythm (N), AF, other abnormal rhythms (O) and noise (P), and it achieved accuracy (acc) 93.08 %, precision (ppv) 78.68 %, sensitivity (sen) 81.84 %, specificity (spec) 94.00 %, and F₁ 0.8382. In addition, it achieved the acc 97.98, ppv 96.40, sen 98.37 %, spec 98.41 %, and F₁ 0.9595 for a synthetic Db consisting of Shandong provincial hospital AF database (SPH AF Db) and CinC 2011 Db for classifying 3 categories i.e., N, AF and P. These results underscore the effectiveness of the ABNet in capturing detailed information about waveform and different frequencies in ECG signals.

为了在低信噪比的心电信号中检测房颤，本研究引入了基于频率分解的自适应帧流网络（ABNet）。ABNet具有显著的优势：它在嘈杂环境中识别AF方面具有很高的鲁棒性，它将心电信号分解为32个频率通道记录以优化频率范围以更好地识别AF，并设计了自适应bin-stream网络以获得最佳结果。该方法利用5级Haar小波包分解将预处理后的心电信号分解为相应的32频通道记录，将预处理后的信号和记录分别送入bin-stream网络的波形流和频率流。最后，采用自适应方法获得最优分类结果。在PhysioNet/Computing in Cardiology Challenge 2017数据库（CinC 2017 Db）中对ABNet进行了验证，对正常窦性心律(N)、AF、其他异常心律(O)和噪声(P) 4类进行了分类，准确率（acc）为93.08%，精密度（ppv）为78.68%，灵敏度（sen）为81.84%，特异性（spec）为94.00%，F1为0.8382。此外，由山东省医院房颤数据库（SPH AF Db）和CinC 2011 Db组成的合成Db对N、AF和p 3个类别进行分类，其准确度为97.98，ppv为96.40，sen为98.37%，spec为98.41%，F1为0.9595，这些结果表明ABNet在捕获心电信号中波形和不同频率的详细信息方面是有效的。

{"title":"An adaptive bin-stream network based on frequency decomposition for classifying atrial fibrillation with low SNR data","authors":"Jilin Wang , Tengqun Shen , Mengfan Li , Yijun Ma , Guozhen Sun , Yatao Zhang","doi":"10.1016/j.medengphy.2025.104412","DOIUrl":"10.1016/j.medengphy.2025.104412","url":null,"abstract":"<div><div>To detect atrial fibrillation (AF) in ECG signals with low signal-to-noise ratio (SNR), this study introduces the adaptive bin-stream network (ABNet) based on frequency decomposition. The ABNet offers notable advantages: it exhibits high robustness in identifying AF amidst noisy environments, it decomposes the ECG signals into 32-frequency channel recordings to refine frequency ranges for better identifying AF, and it designs an adaptive bin-stream network to gain the optimal results. The method utilizes a 5-level Haar wavelet packet decomposition to decompose the preprocessed ECG signals into their corresponding 32-frequency channel recordings, and the preprocessing signals and the recordings are fed into waveform stream and frequency stream of the bin-stream network, respectively. Finally, an adaptive approach is employed to obtain the optimal classification results. The ABNet was validated for the PhysioNet/Computing in Cardiology Challenge 2017 database (CinC 2017 Db) to classify 4 categories i.e., normal sinus rhythm (N), AF, other abnormal rhythms (O) and noise (P), and it achieved accuracy (<em>acc</em>) 93.08 %, precision (<em>ppv</em>) 78.68 %, sensitivity (<em>sen</em>) 81.84 %, specificity (<em>spec</em>) 94.00 %, and <em>F</em><sub>1</sub> 0.8382. In addition, it achieved the <em>acc</em> 97.98, <em>ppv</em> 96.40, <em>sen</em> 98.37 %, <em>spec</em> 98.41 %, and <em>F</em><sub>1</sub> 0.9595 for a synthetic Db consisting of Shandong provincial hospital AF database (SPH AF Db) and CinC 2011 Db for classifying 3 categories i.e., N, AF and P. These results underscore the effectiveness of the ABNet in capturing detailed information about waveform and different frequencies in ECG signals.</div></div>","PeriodicalId":49836,"journal":{"name":"Medical Engineering & Physics","volume":"145 ","pages":"Article 104412"},"PeriodicalIF":2.3,"publicationDate":"2025-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144864385","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}

引用次数: 0

A user-defined element for simulating hydrogel injection into trabecular bone: Numerical simulations and experimental validation 一个用户定义的元素模拟水凝胶注射到小梁骨：数值模拟和实验验证

IF 2.3 4区医学 Q3 ENGINEERING, BIOMEDICAL

Medical Engineering & Physics

Pub Date : 2025-08-05 DOI: 10.1016/j.medengphy.2025.104411

Georgios F. Samaras , Vincent Dischl , Anita Fung , Vincent A. Stadelmann , Ulrike Kettenberger , Stephen J. Ferguson , Benedikt Helgason

In this study, we present a comprehensive numerical model to simulate the injection of hydrogel into femurs. The model is designed to capture the complex interactions between the hydrogel rheological properties and the biomechanical environment of the femur. The coupled mechanical-flow formulation, based on the Theory of Porous Media, is implemented in an open source Abaqus UEL subroutine, where displacements, pressure and saturation are the unknowns. The rheological properties of the hydrogel were calibrated against experimental augmentations in three femurs and the calibrated model was then applied to three different femurs where the hydrogel patterns were compared to experimental data. Furthermore, the simulations demonstrated the effect of injection flow rate and heterogeneous permeability on the hydrogel patterns and quantified the trabecular matrix's solid strains developed during the injection process. The simulations captured well the volume distribution with an average dice coefficient of 0.75 for the three tested specimens. In addition, the calculated solid strains were below the tensile yield limit for the tested flow rate range. A description of the constitutive equations and the implementation into an Abaqus user element subroutine is provided. Overall, our modeling methodology provides a computational tool that can be used to more accurately model bone augmentation and furthermore plan more safely the treatment of osteoporotic patients.

在这项研究中，我们提出了一个综合的数值模型来模拟水凝胶注入股骨。该模型旨在捕捉水凝胶流变特性与股骨生物力学环境之间复杂的相互作用。基于多孔介质理论的耦合力学-流动公式在开源的Abaqus UEL子程序中实现，其中排量、压力和饱和度是未知的。根据三个股骨的实验增强对水凝胶的流变特性进行校准，然后将校准模型应用于三个不同的股骨，并将水凝胶模式与实验数据进行比较。此外，模拟显示了注入流量和非均质渗透率对水凝胶形态的影响，并量化了注入过程中小梁基质的固体应变。模拟结果较好地反映了三种试样的体积分布，平均骰子系数为0.75。此外，计算出的固体应变低于测试流速范围内的拉伸屈服极限。给出了本构方程的描述及其在Abaqus用户元素子程序中的实现。总的来说，我们的建模方法提供了一种计算工具，可以用来更准确地模拟骨增强，并进一步计划更安全的骨质疏松症患者的治疗。

{"title":"A user-defined element for simulating hydrogel injection into trabecular bone: Numerical simulations and experimental validation","authors":"Georgios F. Samaras , Vincent Dischl , Anita Fung , Vincent A. Stadelmann , Ulrike Kettenberger , Stephen J. Ferguson , Benedikt Helgason","doi":"10.1016/j.medengphy.2025.104411","DOIUrl":"10.1016/j.medengphy.2025.104411","url":null,"abstract":"<div><div>In this study, we present a comprehensive numerical model to simulate the injection of hydrogel into femurs. The model is designed to capture the complex interactions between the hydrogel rheological properties and the biomechanical environment of the femur. The coupled mechanical-flow formulation, based on the Theory of Porous Media, is implemented in an open source Abaqus UEL subroutine, where displacements, pressure and saturation are the unknowns. The rheological properties of the hydrogel were calibrated against experimental augmentations in three femurs and the calibrated model was then applied to three different femurs where the hydrogel patterns were compared to experimental data. Furthermore, the simulations demonstrated the effect of injection flow rate and heterogeneous permeability on the hydrogel patterns and quantified the trabecular matrix's solid strains developed during the injection process. The simulations captured well the volume distribution with an average dice coefficient of 0.75 for the three tested specimens. In addition, the calculated solid strains were below the tensile yield limit for the tested flow rate range. A description of the constitutive equations and the implementation into an Abaqus user element subroutine is provided. Overall, our modeling methodology provides a computational tool that can be used to more accurately model bone augmentation and furthermore plan more safely the treatment of osteoporotic patients.</div></div>","PeriodicalId":49836,"journal":{"name":"Medical Engineering & Physics","volume":"145 ","pages":"Article 104411"},"PeriodicalIF":2.3,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144841540","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}

引用次数: 0

Fast geometric deep learning for intraoperative soft tissue deformation estimation: Towards real-time AR guidance in liver surgery 快速几何深度学习用于术中软组织变形估计：面向肝脏手术实时AR引导

IF 2.3 4区医学 Q3 ENGINEERING, BIOMEDICAL

Medical Engineering & Physics

Pub Date : 2025-08-05 DOI: 10.1016/j.medengphy.2025.104409

Zixuan Zhai , Enpeng Wang , Xiaojun Chen

The real-time computation of the intraoperative spatial positioning of soft tissues, particularly those not visible within the body, such as blood vessels, is crucial for augmented reality navigation systems. Conventional biomechanical models face challenges in real-time computation and the acquisition of boundary conditions. A novel deep learning framework is proposed, integrating an optimized PointNet++ architecture for modelling liver and vascular deformation. The framework utilizes multi-scale feature extraction, lightweight self-attention mechanisms, and residual feature propagation to predict vascular displacement fields and normal vectors. A hybrid loss function that integrates Chamfer distance and MSE losses improves geometric consistency and deformation accuracy. The proposed approach, utilizing finite element method (FEM)-simulated datasets of liver stretching procedures, exhibits enhanced performance with root mean square errors (RMSE) of 2.78 ± 0.69 mm for hepatic veins and 1.81 ± 0.74 mm for portal veins. This method surpasses conventional techniques by 37.5% in accuracy and reduces inference time to 0.25 seconds. The optimized network exhibits a computation speed that is 83.9% faster than leading non-rigid registration algorithms. Subsequent tumour localization experiments demonstrate a targeting accuracy of 3.2 mm via vascular topology analysis, confirming clinical relevance. This research develops an effective framework for predicting deformation in real-time, providing a significant advancement for navigation in AR-guided hepatobiliary surgery.

术中软组织空间定位的实时计算，特别是那些在体内不可见的软组织，如血管，对于增强现实导航系统至关重要。传统的生物力学模型在实时计算和边界条件获取方面面临挑战。提出了一种新的深度学习框架，集成了优化的PointNet++架构，用于肝脏和血管变形建模。该框架利用多尺度特征提取、轻量级自关注机制和残差特征传播来预测血管位移场和法向量。混合损失函数集成了倒角距离和MSE损失，提高了几何一致性和变形精度。该方法利用有限元法（FEM）模拟肝脏拉伸过程的数据集，显示出增强的性能，肝静脉的均方根误差（RMSE）为2.78±0.69 mm，门静脉的均方根误差为1.81±0.74 mm。该方法的准确率比传统方法提高了37.5%，推理时间缩短到0.25秒。优化后的网络计算速度比现有的非刚性配准算法快83.9%。随后的肿瘤定位实验表明，通过血管拓扑分析，靶向精度为3.2毫米，证实了临床相关性。本研究开发了一种实时预测变形的有效框架，为ar引导的肝胆手术导航提供了重大进展。

{"title":"Fast geometric deep learning for intraoperative soft tissue deformation estimation: Towards real-time AR guidance in liver surgery","authors":"Zixuan Zhai , Enpeng Wang , Xiaojun Chen","doi":"10.1016/j.medengphy.2025.104409","DOIUrl":"10.1016/j.medengphy.2025.104409","url":null,"abstract":"<div><div>The real-time computation of the intraoperative spatial positioning of soft tissues, particularly those not visible within the body, such as blood vessels, is crucial for augmented reality navigation systems. Conventional biomechanical models face challenges in real-time computation and the acquisition of boundary conditions. A novel deep learning framework is proposed, integrating an optimized PointNet++ architecture for modelling liver and vascular deformation. The framework utilizes multi-scale feature extraction, lightweight self-attention mechanisms, and residual feature propagation to predict vascular displacement fields and normal vectors. A hybrid loss function that integrates Chamfer distance and MSE losses improves geometric consistency and deformation accuracy. The proposed approach, utilizing finite element method (FEM)-simulated datasets of liver stretching procedures, exhibits enhanced performance with root mean square errors (RMSE) of 2.78 ± 0.69 mm for hepatic veins and 1.81 ± 0.74 mm for portal veins. This method surpasses conventional techniques by 37.5% in accuracy and reduces inference time to 0.25 seconds. The optimized network exhibits a computation speed that is 83.9% faster than leading non-rigid registration algorithms. Subsequent tumour localization experiments demonstrate a targeting accuracy of 3.2 mm via vascular topology analysis, confirming clinical relevance. This research develops an effective framework for predicting deformation in real-time, providing a significant advancement for navigation in AR-guided hepatobiliary surgery.</div></div>","PeriodicalId":49836,"journal":{"name":"Medical Engineering & Physics","volume":"145 ","pages":"Article 104409"},"PeriodicalIF":2.3,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144809762","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}

引用次数: 0

Comparative analysis of Electrospun PLA fibers incorporating bioactive glass nanoparticles: morphological, biological, and osteogenic properties for bone regeneration 含有生物活性玻璃纳米颗粒的静电纺PLA纤维的比较分析：形态学、生物学和骨再生的成骨特性

IF 2.3 4区医学 Q3 ENGINEERING, BIOMEDICAL

Medical Engineering & Physics

Pub Date : 2025-07-31 DOI: 10.1016/j.medengphy.2025.104410

Brunna da Silva Nobrega Souza , Lilian de Siqueira , Marina Santos Fernandes , Joyce Rodrigues de Souza , Elisa Camargo Kukulka , Letícia Adrielly Dias Grisante , Tiago Moreira Bastos Campos , Luana Marotta Reis de Vasconcellos , Alexandre Luiz Souto Borges

Polylactic acid (PLA) is widely studied for bone repair due to its biodegradability, biocompatibility, and bioresorbability. However, its limited bioactivity and hydrophobic surface hinder optimal cell interaction and integration. Incorporating bioactive glass (BG) particles into PLA scaffolds via electrospinning and electrospray techniques has emerged as a promising strategy to improve biological performance. This study aimed to fabricate and characterize PLA scaffolds, both with incorporated and surface-coated BG, and to assess their osteogenic potential for tissue engineering applications. Scaffold morphology was evaluated by scanning electron microscopy, and biological performance was assessed through in vitro assays using mesenchymal stem cells derived from Wistar rat bone marrow. Cell viability, total protein content, alkaline phosphatase (ALP) activity, and mineralized nodule formation were analyzed. The scaffolds displayed porous, interconnected structures with fiber diameters influenced by BG incorporation method. All groups demonstrated cytocompatibility, while scaffolds containing BG both incorporated and sprayed—showed significantly higher ALP activity, suggesting enhanced osteogenic differentiation. Mineralization nodules further confirmed the induction of osteogenesis. These findings highlight the potential of PLA/BG composite scaffolds, especially when functionalized via combined electrospinning and electrospray methods, as a promising platform for bone tissue engineering.

聚乳酸（PLA）由于其生物可降解性、生物相容性和生物可吸收性而被广泛研究用于骨修复。然而，其有限的生物活性和疏水表面阻碍了最佳的细胞相互作用和整合。通过静电纺丝和电喷雾技术将生物活性玻璃（BG）颗粒掺入聚乳酸支架中是一种很有前途的提高生物性能的策略。本研究旨在制备和表征PLA支架，包括掺入和表面包覆的BG，并评估其在组织工程中的成骨潜力。利用扫描电镜观察支架形态，利用Wistar大鼠骨髓间充质干细胞体外检测支架生物学性能。分析细胞活力、总蛋白含量、碱性磷酸酶（ALP）活性和矿化结节形成情况。受BG掺入法影响，支架呈现出多孔、互联的结构，纤维直径受到影响。所有组均表现出细胞相容性，而含有BG的支架（包括掺入和喷雾）均表现出更高的ALP活性，表明成骨分化增强。矿化结节进一步证实了骨生成的诱导作用。这些发现突出了PLA/BG复合支架的潜力，特别是当通过静电纺丝和电喷雾相结合的方法实现功能化时，作为骨组织工程的一个有前途的平台。

{"title":"Comparative analysis of Electrospun PLA fibers incorporating bioactive glass nanoparticles: morphological, biological, and osteogenic properties for bone regeneration","authors":"Brunna da Silva Nobrega Souza , Lilian de Siqueira , Marina Santos Fernandes , Joyce Rodrigues de Souza , Elisa Camargo Kukulka , Letícia Adrielly Dias Grisante , Tiago Moreira Bastos Campos , Luana Marotta Reis de Vasconcellos , Alexandre Luiz Souto Borges","doi":"10.1016/j.medengphy.2025.104410","DOIUrl":"10.1016/j.medengphy.2025.104410","url":null,"abstract":"<div><div>Polylactic acid (PLA) is widely studied for bone repair due to its biodegradability, biocompatibility, and bioresorbability. However, its limited bioactivity and hydrophobic surface hinder optimal cell interaction and integration. Incorporating bioactive glass (BG) particles into PLA scaffolds via electrospinning and electrospray techniques has emerged as a promising strategy to improve biological performance. This study aimed to fabricate and characterize PLA scaffolds, both with incorporated and surface-coated BG, and to assess their osteogenic potential for tissue engineering applications. Scaffold morphology was evaluated by scanning electron microscopy, and biological performance was assessed through in vitro assays using mesenchymal stem cells derived from Wistar rat bone marrow. Cell viability, total protein content, alkaline phosphatase (ALP) activity, and mineralized nodule formation were analyzed. The scaffolds displayed porous, interconnected structures with fiber diameters influenced by BG incorporation method. All groups demonstrated cytocompatibility, while scaffolds containing BG both incorporated and sprayed—showed significantly higher ALP activity, suggesting enhanced osteogenic differentiation. Mineralization nodules further confirmed the induction of osteogenesis. These findings highlight the potential of PLA/BG composite scaffolds, especially when functionalized via combined electrospinning and electrospray methods, as a promising platform for bone tissue engineering.</div></div>","PeriodicalId":49836,"journal":{"name":"Medical Engineering & Physics","volume":"145 ","pages":"Article 104410"},"PeriodicalIF":2.3,"publicationDate":"2025-07-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144908034","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}

引用次数: 0

Reduced model aided fluid-structure interaction design framework for shunt systems 简化模型辅助分流系统流固交互设计框架

IF 2.3 4区医学 Q3 ENGINEERING, BIOMEDICAL

Medical Engineering & Physics

Pub Date : 2025-07-28 DOI: 10.1016/j.medengphy.2025.104403

Elizabeth Hayman , Van Dung Nguyen , Ian S. McFarlane , Juliette Pech , Jayaratnam Jayamohan , José-Maria Peña Sánchez , Sarah Waters , Antoine Jerusalem

Traditionally, clinical devices are designed, tested and improved through lengthy and expensive laboratory experiments and clinical trials [1]. More recently, computational methods have allowed for rapid testing, speeding up the design process and enabling far more complete searches of design space. While computational models cannot fully capture the complexities of biological systems, they provide valuable insights into crucial underlying mechanisms, such as the effects of fluid-structure interactions (FSIs). In this paper we present a modular, partitioned, computational FSI pipeline whereby 2D reduced order models guide the 3D design of the problem of interest. This framework is applied to the problem of hydrocephalus shunt occlusion. Hydrocephalus is a medical condition characterised by an excess of cerebrospinal fluid (CSF) in the brain, and is commonly treated with the insertion of a shunt system. This system includes a ventricular catheter component – a hollow tube with inlet holes arranged in the tube wall close to the closed tip – which is positioned in the lateral ventricles of the brain. Despite recent improvements in the catheter material, this treatment still has high failure rates, most often due to the blockage of the catheter by the Choroid Plexus (ChP) tissue. We use an idealised FSI model to compare existing catheter designs by considering the deformation of the ChP under CSF flow in the ventricle environment in an hydrocephalus scenario. To the best of our knowledge, this is the first computational framework to directly incorporate the deformation of the ChP to discriminate between catheter designs. The faster 2D model is used in a comprehensive parameter sweep of the catheter design domain, and motivates a new design, then confirmed to be an improvement when tested in the full 3D domain. This approach demonstrates the success of using reduced order methods to guide the design of a more complex problem.

传统上，临床设备是通过漫长而昂贵的实验室实验和临床试验来设计、测试和改进的。最近，计算方法允许快速测试，加速设计过程，并使更完整的设计空间搜索成为可能。虽然计算模型不能完全捕捉生物系统的复杂性，但它们为关键的潜在机制提供了有价值的见解，例如流固相互作用（FSIs）的影响。在本文中，我们提出了一个模块化，分区，计算FSI管道，其中二维降阶模型指导感兴趣的问题的三维设计。该框架适用于脑积水分流闭塞的问题。脑积水是一种以脑脊液（CSF）过多为特征的医学病症，通常通过插入分流系统进行治疗。该系统包括一个心室导管组件——一个位于大脑侧脑室的空心管，其入口孔位于靠近封闭尖端的管壁上。尽管最近导管材料有所改进，但这种治疗仍然有很高的失败率，最常见的原因是脉络膜丛（ChP）组织阻塞导管。我们使用一个理想化的FSI模型来比较现有的导管设计，通过考虑脑积水情况下脑室环境中脑脊液流下ChP的变形。据我们所知，这是第一个直接结合ChP变形来区分导管设计的计算框架。更快的2D模型用于导管设计领域的综合参数扫描，并激发新的设计，然后在全3D领域进行测试时确认是一种改进。这种方法证明了使用降阶方法来指导更复杂问题的设计的成功。

{"title":"Reduced model aided fluid-structure interaction design framework for shunt systems","authors":"Elizabeth Hayman , Van Dung Nguyen , Ian S. McFarlane , Juliette Pech , Jayaratnam Jayamohan , José-Maria Peña Sánchez , Sarah Waters , Antoine Jerusalem","doi":"10.1016/j.medengphy.2025.104403","DOIUrl":"10.1016/j.medengphy.2025.104403","url":null,"abstract":"<div><div>Traditionally, clinical devices are designed, tested and improved through lengthy and expensive laboratory experiments and clinical trials <span><span>[1]</span></span>. More recently, computational methods have allowed for rapid testing, speeding up the design process and enabling far more complete searches of design space. While computational models cannot fully capture the complexities of biological systems, they provide valuable insights into crucial underlying mechanisms, such as the effects of fluid-structure interactions (FSIs). In this paper we present a modular, partitioned, computational FSI pipeline whereby 2D reduced order models guide the 3D design of the problem of interest. This framework is applied to the problem of hydrocephalus shunt occlusion. Hydrocephalus is a medical condition characterised by an excess of cerebrospinal fluid (CSF) in the brain, and is commonly treated with the insertion of a shunt system. This system includes a ventricular catheter component – a hollow tube with inlet holes arranged in the tube wall close to the closed tip – which is positioned in the lateral ventricles of the brain. Despite recent improvements in the catheter material, this treatment still has high failure rates, most often due to the blockage of the catheter by the Choroid Plexus (ChP) tissue. We use an idealised FSI model to compare existing catheter designs by considering the deformation of the ChP under CSF flow in the ventricle environment in an hydrocephalus scenario. To the best of our knowledge, this is the first computational framework to directly incorporate the deformation of the ChP to discriminate between catheter designs. The faster 2D model is used in a comprehensive parameter sweep of the catheter design domain, and motivates a new design, then confirmed to be an improvement when tested in the full 3D domain. This approach demonstrates the success of using reduced order methods to guide the design of a more complex problem.</div></div>","PeriodicalId":49836,"journal":{"name":"Medical Engineering & Physics","volume":"144 ","pages":"Article 104403"},"PeriodicalIF":2.3,"publicationDate":"2025-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144772749","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}

引用次数: 0

A numerical investigation of the kinematic and fluid dynamic behaviour of an intramuscular autoinjector designed for optimising injection efficiency 为优化注射效率而设计的肌肉内自动注射器的运动学和流体动力学行为的数值研究

IF 2.3 4区医学 Q3 ENGINEERING, BIOMEDICAL

Medical Engineering & Physics

Pub Date : 2025-07-26 DOI: 10.1016/j.medengphy.2025.104407

Sudesh Sivarasu , Ntokozo Magubane , Chibuike Mbanefo , Malebogo Ngoepe

The usability and versatility of autoinjectors in managing chronic and autoimmune diseases have made them increasingly attractive in medicine. However, investigations into autoinjector designs require an understanding of the kinematic properties and fluid behaviour during injection. To optimise injection efficiency, this study develops a mathematical and computational fluid dynamics (CFD) model of an IM autoinjector by investigating the effects of viscosity, needle length, needle diameter, and medication volume on the injection process. The model was verified and validated using a comparator experiment and optimised using a parameter sensitivity analysis. The mathematical model results show plunger displacement increases linearly in low viscous fluids (v < 20 cP), allowing faster injections. CFD simulations show that high-viscosity fluids (v > 20 cP) reduce injectability and increase syringeability. Needle gauges below 20 exhibited constant dynamic pressure and negligible shear stress, while gauges between 20 and 25 showed higher shear stress and pressure variability. Longer needles and larger medication volumes increase dynamic pressure and shear stress, prolonging injection time. The mathematical and CFD models matched experimental measurements within a 1.1 % and 4.8 % margin of error, respectively. These findings inform the design of efficient autoinjectors, enhancing drug delivery, patient comfort, and compliance.

自体注射器在治疗慢性和自身免疫性疾病方面的可用性和多功能性使其在医学上越来越有吸引力。然而，对自动进样器设计的研究需要了解注入过程中的运动特性和流体行为。为了优化注射效率，本研究通过研究粘度、针头长度、针头直径和药物体积对注射过程的影响，建立了IM自动注射器的数学和计算流体动力学（CFD）模型。采用比较器实验对模型进行了验证和验证，并采用参数敏感性分析对模型进行了优化。数学模型结果表明，柱塞位移在低粘性流体中呈线性增加(v <；20cp)，从而加快注射速度。CFD模拟表明，高粘度流体(v >；20cp)降低可注射性，增加可注射性。针规在20以下表现出恒定的动压力和可忽略的剪切应力，而在20和25之间的针规表现出较高的剪切应力和压力变异性。更长的针头和更大的药量增加了动压力和剪切应力，延长了注射时间。数学模型和CFD模型与实验测量值的匹配误差分别在1.1%和4.8%以内。这些发现为设计高效的自体注射器提供了信息，增强了药物输送、患者舒适度和依从性。

{"title":"A numerical investigation of the kinematic and fluid dynamic behaviour of an intramuscular autoinjector designed for optimising injection efficiency","authors":"Sudesh Sivarasu , Ntokozo Magubane , Chibuike Mbanefo , Malebogo Ngoepe","doi":"10.1016/j.medengphy.2025.104407","DOIUrl":"10.1016/j.medengphy.2025.104407","url":null,"abstract":"<div><div>The usability and versatility of autoinjectors in managing chronic and autoimmune diseases have made them increasingly attractive in medicine. However, investigations into autoinjector designs require an understanding of the kinematic properties and fluid behaviour during injection. To optimise injection efficiency, this study develops a mathematical and computational fluid dynamics (CFD) model of an IM autoinjector by investigating the effects of viscosity, needle length, needle diameter, and medication volume on the injection process. The model was verified and validated using a comparator experiment and optimised using a parameter sensitivity analysis. The mathematical model results show plunger displacement increases linearly in low viscous fluids (<em>v</em> < 20 cP), allowing faster injections. CFD simulations show that high-viscosity fluids (<em>v</em> > 20 cP) reduce injectability and increase syringeability. Needle gauges below 20 exhibited constant dynamic pressure and negligible shear stress, while gauges between 20 and 25 showed higher shear stress and pressure variability. Longer needles and larger medication volumes increase dynamic pressure and shear stress, prolonging injection time. The mathematical and CFD models matched experimental measurements within a 1.1 % and 4.8 % margin of error, respectively. These findings inform the design of efficient autoinjectors, enhancing drug delivery, patient comfort, and compliance.</div></div>","PeriodicalId":49836,"journal":{"name":"Medical Engineering & Physics","volume":"144 ","pages":"Article 104407"},"PeriodicalIF":2.3,"publicationDate":"2025-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144739425","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}

引用次数: 0

Explainable deep learning framework for brain tumor detection: Integrating LIME, Grad-CAM, and SHAP for enhanced accuracy 用于脑肿瘤检测的可解释的深度学习框架：整合LIME， Grad-CAM和SHAP以提高准确性

IF 2.3 4区医学 Q3 ENGINEERING, BIOMEDICAL

Medical Engineering & Physics

Pub Date : 2025-07-25 DOI: 10.1016/j.medengphy.2025.104405

Abdurrahim Akgündoğdu , Şerife Çelikbaş

Deep learning approaches have improved disease diagnosis efficiency. However, AI-based decision systems lack sufficient transparency and interpretability. This study aims to enhance the explainability and training performance of deep learning models using explainable artificial intelligence (XAI) techniques for brain tumor detection. A two-stage training approach and XAI methods were implemented. The proposed convolutional neural network achieved 97.20% accuracy, 98.00% sensitivity, 96.40% specificity, and 98.90% ROC-AUC on the BRATS2019 dataset. It was analyzed with explainability techniques including Local Interpretable Model-Agnostic Explanations (LIME), Gradient-weighted Class Activation Mapping (Grad-CAM), and Shapley Additive Explanations (SHAP). The masks generated from these analyses enhanced the dataset, leading to a higher accuracy of 99.40%, 99.20% sensitivity, 99.60% specificity, 99.60% precision, and 99.90% ROC-AUC in the final stage. The integration of LIME, Grad-CAM, and SHAP showed significant success by increasing the accuracy performance of the model from 97.20% to 99.40%. Furthermore, the model was evaluated for fidelity, stability, and consistency and showed reliable and stable results. The same strategy was applied to the BR35H dataset to test the generalizability of the model, and the accuracy increased from 96.80% to 99.80% on this dataset as well, supporting the effectiveness of the method on different data sources.

深度学习方法提高了疾病诊断效率。然而，基于人工智能的决策系统缺乏足够的透明度和可解释性。本研究旨在利用可解释人工智能（XAI）技术增强深度学习模型的可解释性和训练性能，用于脑肿瘤检测。采用两阶段训练方法和XAI方法。本文提出的卷积神经网络在BRATS2019数据集上的准确率为97.20%，灵敏度为98.00%，特异性为96.40%，ROC-AUC为98.90%。采用局部可解释模型不可知解释（LIME）、梯度加权类激活映射（Grad-CAM）和Shapley加性解释（SHAP）等可解释性技术对其进行分析。从这些分析中生成的掩模增强了数据集，在最后阶段达到99.40%的准确率、99.20%的灵敏度、99.60%的特异性、99.60%的精度和99.90%的ROC-AUC。LIME、Grad-CAM和SHAP的整合取得了显著的成功，将模型的准确率从97.20%提高到99.40%。对模型进行保真度、稳定性和一致性评价，结果可靠稳定。将相同的策略应用于BR35H数据集，测试模型的泛化性，该数据集的准确率也从96.80%提高到99.80%，支持了该方法在不同数据源上的有效性。

{"title":"Explainable deep learning framework for brain tumor detection: Integrating LIME, Grad-CAM, and SHAP for enhanced accuracy","authors":"Abdurrahim Akgündoğdu , Şerife Çelikbaş","doi":"10.1016/j.medengphy.2025.104405","DOIUrl":"10.1016/j.medengphy.2025.104405","url":null,"abstract":"<div><div>Deep learning approaches have improved disease diagnosis efficiency. However, AI-based decision systems lack sufficient transparency and interpretability. This study aims to enhance the explainability and training performance of deep learning models using explainable artificial intelligence (XAI) techniques for brain tumor detection. A two-stage training approach and XAI methods were implemented. The proposed convolutional neural network achieved 97.20% accuracy, 98.00% sensitivity, 96.40% specificity, and 98.90% ROC-AUC on the BRATS2019 dataset. It was analyzed with explainability techniques including Local Interpretable Model-Agnostic Explanations (LIME), Gradient-weighted Class Activation Mapping (Grad-CAM), and Shapley Additive Explanations (SHAP). The masks generated from these analyses enhanced the dataset, leading to a higher accuracy of 99.40%, 99.20% sensitivity, 99.60% specificity, 99.60% precision, and 99.90% ROC-AUC in the final stage. The integration of LIME, Grad-CAM, and SHAP showed significant success by increasing the accuracy performance of the model from 97.20% to 99.40%. Furthermore, the model was evaluated for fidelity, stability, and consistency and showed reliable and stable results. The same strategy was applied to the BR35H dataset to test the generalizability of the model, and the accuracy increased from 96.80% to 99.80% on this dataset as well, supporting the effectiveness of the method on different data sources.</div></div>","PeriodicalId":49836,"journal":{"name":"Medical Engineering & Physics","volume":"144 ","pages":"Article 104405"},"PeriodicalIF":2.3,"publicationDate":"2025-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144723271","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}

引用次数: 0

A machine learning approach to concussive group classification using discrete outcome measures from a low-cost movement-based assessment system 一种机器学习方法，使用来自低成本的基于运动的评估系统的离散结果测量进行震荡组分类

IF 2.3 4区医学 Q3 ENGINEERING, BIOMEDICAL

Medical Engineering & Physics

Pub Date : 2025-07-24 DOI: 10.1016/j.medengphy.2025.104402

Jacob M. Thomas , Jamie B. Hall , Rebecca Bliss , Emily Leary , Stephen P. Sayers , Praveen Rao , Trent M. Guess

Measurable neuromotor control deficits during functional task performance could provide objective criteria to aid in concussion diagnosis. However, many tools which measure these constructs are unidimensional and not clinically feasible. The purpose of this study was to assess the classification accuracy of a machine learning model using features measured by a clinically feasible movement-based assessment system (Mizzou Point-of-care Assessment System (MPASS) between athletes with and without concussion. Forty collegiate athletes participated. Twenty (19.40 ± 1.04 yrs., 11 females) suffered concussion within two weeks of data collection (5.40 ± 3.68 days). Twenty (19.85 ± 1.20 yrs.) sex, sport, and position-matched athletes had no concussions in the past year. All participants completed three 30-second static balance trials with eyes closed on foam surface under both single task and cognitive dual task conditions, four trials of gait under normal, head shaking, and dual task conditions, and reaction time tasks. Kinematics, kinetics, and reaction times were recorded by MPASS. Measures were used as features for a XGBoost machine learning model. Five-fold cross-validation yielded mean (across 5-folds): 82.5 % accuracy, 75 % sensitivity, 90 % specificity, 88.2 % positive predictive value, and 78.3 % negative predictive value. Results indicate promise for using movement-based features from a low-cost movement-based assessment system to improve the objectivity of concussion diagnosis decision-making.

在功能性任务表现中可测量的神经运动控制缺陷可为脑震荡诊断提供客观标准。然而，许多测量这些结构的工具是单向度的，在临床上不可行。本研究的目的是评估机器学习模型的分类准确性，使用临床可行的基于运动的评估系统（Mizzou Point-of-care assessment system， MPASS）在有脑震荡和没有脑震荡的运动员之间测量的特征。40名大学生运动员参加了比赛。20（19.40±1.04）年。11例女性)在收集数据的2周内（5.40±3.68天）出现脑震荡。20名（19.85±1.20岁）性别、运动和位置匹配的运动员在过去一年中没有发生过脑震荡。所有被试分别在单任务和认知双任务条件下完成3个30秒闭眼泡沫表面静态平衡试验，在正常、摇头和双任务条件下完成4个步态试验，以及反应时间任务。通过MPASS记录运动学、动力学和反应时间。度量被用作XGBoost机器学习模型的特征。5倍交叉验证的平均准确率为82.5%，灵敏度为75%，特异性为90%，阳性预测值为88.2%，阴性预测值为78.3%。结果表明，利用基于运动特征的低成本评估系统可以提高脑震荡诊断决策的客观性。

{"title":"A machine learning approach to concussive group classification using discrete outcome measures from a low-cost movement-based assessment system","authors":"Jacob M. Thomas , Jamie B. Hall , Rebecca Bliss , Emily Leary , Stephen P. Sayers , Praveen Rao , Trent M. Guess","doi":"10.1016/j.medengphy.2025.104402","DOIUrl":"10.1016/j.medengphy.2025.104402","url":null,"abstract":"<div><div>Measurable neuromotor control deficits during functional task performance could provide objective criteria to aid in concussion diagnosis. However, many tools which measure these constructs are unidimensional and not clinically feasible. The purpose of this study was to assess the classification accuracy of a machine learning model using features measured by a clinically feasible movement-based assessment system (Mizzou Point-of-care Assessment System (MPASS) between athletes with and without concussion. Forty collegiate athletes participated. Twenty (19.40 ± 1.04 yrs., 11 females) suffered concussion within two weeks of data collection (5.40 ± 3.68 days). Twenty (19.85 ± 1.20 yrs.) sex, sport, and position-matched athletes had no concussions in the past year. All participants completed three 30-second static balance trials with eyes closed on foam surface under both single task and cognitive dual task conditions, four trials of gait under normal, head shaking, and dual task conditions, and reaction time tasks. Kinematics, kinetics, and reaction times were recorded by MPASS. Measures were used as features for a XGBoost machine learning model. Five-fold cross-validation yielded mean (across 5-folds): 82.5 % accuracy, 75 % sensitivity, 90 % specificity, 88.2 % positive predictive value, and 78.3 % negative predictive value. Results indicate promise for using movement-based features from a low-cost movement-based assessment system to improve the objectivity of concussion diagnosis decision-making.</div></div>","PeriodicalId":49836,"journal":{"name":"Medical Engineering & Physics","volume":"144 ","pages":"Article 104402"},"PeriodicalIF":2.3,"publicationDate":"2025-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144772750","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}

引用次数: 0

TPC-GCN: Deep learning for pulse pattern classification in traditional Chinese medicine TPC-GCN：中医脉象分类的深度学习

IF 1.7 4区医学 Q3 ENGINEERING, BIOMEDICAL

Medical Engineering & Physics

Pub Date : 2025-07-23 DOI: 10.1016/j.medengphy.2025.104401

Hui Li , Yuetang Li , Zhidong Zhang , Chenyang Xue , Zhenhua Li , Xiaobo Li , Jiuzhang Men

Pulse diagnosis holds a pivotal role in traditional Chinese medicine (TCM) diagnostics, with pulse characteristics serving as one of the critical bases for its assessment. Accurate classification of these pulse pattern is paramount for the objectification of TCM. This study proposes an enhanced SMOTE approach to achieve data augmentation, followed by multi-domain feature extraction. Graph data structures with varying configurations are subsequently constructed to facilitate more profound insights into the intrinsic information within the data. Additionally, a multi-channel lightweight graph convolutional network (GCN) is devised. This network's core strategy lies in extracting diverse layers of information through parallel branches, integrating local structural information with adaptive weights, and employing attention-weighted fusion to improve classification accuracy and model robustness. The proposed network model achieved 91.68% accuracy, a mean F1 score of 92%, a mean recall rate of 92%, and a mean precision rate of 92% on the pulse dataset. The results demonstrate a marked improvement in pulse classification accuracy, validating the efficacy of this approach while offering new perspectives and methodologies for pulse signal classification research.

脉象诊断在中医诊断中占有举足轻重的地位，脉象特征是评价脉象优劣的重要依据之一。这些脉象的准确分类对于中医的客观化至关重要。本研究提出了一种增强的SMOTE方法来实现数据增强，然后进行多域特征提取。随后构建具有不同配置的图数据结构，以便更深入地了解数据中的内在信息。此外，还设计了一个多通道轻量级图卷积网络（GCN）。该网络的核心策略是通过并行分支提取不同层次的信息，将局部结构信息与自适应权值相结合，并采用注意力加权融合来提高分类精度和模型鲁棒性。该网络模型在脉冲数据集上的准确率为91.68%，平均F1得分为92%，平均查全率为92%，平均查准率为92%。结果表明，脉冲信号分类精度显著提高，验证了该方法的有效性，同时为脉冲信号分类研究提供了新的视角和方法。

{"title":"TPC-GCN: Deep learning for pulse pattern classification in traditional Chinese medicine","authors":"Hui Li , Yuetang Li , Zhidong Zhang , Chenyang Xue , Zhenhua Li , Xiaobo Li , Jiuzhang Men","doi":"10.1016/j.medengphy.2025.104401","DOIUrl":"10.1016/j.medengphy.2025.104401","url":null,"abstract":"<div><div>Pulse diagnosis holds a pivotal role in traditional Chinese medicine (TCM) diagnostics, with pulse characteristics serving as one of the critical bases for its assessment. Accurate classification of these pulse pattern is paramount for the objectification of TCM. This study proposes an enhanced SMOTE approach to achieve data augmentation, followed by multi-domain feature extraction. Graph data structures with varying configurations are subsequently constructed to facilitate more profound insights into the intrinsic information within the data. Additionally, a multi-channel lightweight graph convolutional network (GCN) is devised. This network's core strategy lies in extracting diverse layers of information through parallel branches, integrating local structural information with adaptive weights, and employing attention-weighted fusion to improve classification accuracy and model robustness. The proposed network model achieved 91.68% accuracy, a mean F1 score of 92%, a mean recall rate of 92%, and a mean precision rate of 92% on the pulse dataset. The results demonstrate a marked improvement in pulse classification accuracy, validating the efficacy of this approach while offering new perspectives and methodologies for pulse signal classification research.</div></div>","PeriodicalId":49836,"journal":{"name":"Medical Engineering & Physics","volume":"144 ","pages":"Article 104401"},"PeriodicalIF":1.7,"publicationDate":"2025-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144696800","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}

引用次数: 0

A cascade approach for the early detection and localization of myocardial infarction in 2D-echocardiography 二维超声心动图对心肌梗死早期检测和定位的级联方法。

IF 2.3 4区医学 Q3 ENGINEERING, BIOMEDICAL

Medical Engineering & Physics

Pub Date : 2025-07-17 DOI: 10.1016/j.medengphy.2025.104400

Carolina Gomez , Annalisa Letizia , Vincenza Tufano , Filippo Molinari , Massimo Salvi

Myocardial infarction (MI) detection and localization through echocardiography are crucial for effective patient management. However, current diagnostic approaches rely heavily on visual assessment, which can be subjective. In this work we developed a cascade framework for automated MI diagnosis and localization in echocardiograms. Our method combines deep learning for left ventricle wall segmentation with machine learning classification using clinically relevant features. Specifically, we employ a U-Net architecture for segmentation, followed by a two-stage Random Forest classifier for MI detection and localization. We trained and evaluated our approach on two public datasets – CAMUS and HMC-QU. The proposed method achieved 100 % sensitivity and 89.8 % specificity for segment identification, outperforming single-stage classification methods. To the best of our knowledge, this is the first study to apply a multi-step artificial intelligence system combining segmentation and classification for MI diagnosis from echocardiography. This interpretable cascade framework exhibits high performance for early detection and localization of myocardial infarction, demonstrating potential as a clinical decision support tool.

通过超声心动图检测和定位心肌梗死（MI）对有效的患者管理至关重要。然而，目前的诊断方法严重依赖于视觉评估，这可能是主观的。在这项工作中，我们开发了一个级联框架，用于在超声心动图中自动诊断和定位心肌梗死。我们的方法结合了左心室壁分割的深度学习和使用临床相关特征的机器学习分类。具体来说，我们使用U-Net架构进行分割，然后使用两阶段随机森林分类器进行MI检测和定位。我们在CAMUS和HMC-QU两个公共数据集上训练和评估了我们的方法。该方法的灵敏度为100%，特异性为89.8%，优于单阶段分类方法。据我们所知，这是第一个将多步骤人工智能系统结合分割和分类用于超声心动图诊断心肌梗死的研究。这种可解释的级联框架在心肌梗死的早期检测和定位方面表现出高性能，显示出作为临床决策支持工具的潜力。

{"title":"A cascade approach for the early detection and localization of myocardial infarction in 2D-echocardiography","authors":"Carolina Gomez , Annalisa Letizia , Vincenza Tufano , Filippo Molinari , Massimo Salvi","doi":"10.1016/j.medengphy.2025.104400","DOIUrl":"10.1016/j.medengphy.2025.104400","url":null,"abstract":"<div><div>Myocardial infarction (MI) detection and localization through echocardiography are crucial for effective patient management. However, current diagnostic approaches rely heavily on visual assessment, which can be subjective. In this work we developed a cascade framework for automated MI diagnosis and localization in echocardiograms. Our method combines deep learning for left ventricle wall segmentation with machine learning classification using clinically relevant features. Specifically, we employ a U-Net architecture for segmentation, followed by a two-stage Random Forest classifier for MI detection and localization. We trained and evaluated our approach on two public datasets – CAMUS and HMC-QU. The proposed method achieved 100 % sensitivity and 89.8 % specificity for segment identification, outperforming single-stage classification methods. To the best of our knowledge, this is the first study to apply a multi-step artificial intelligence system combining segmentation and classification for MI diagnosis from echocardiography. This interpretable cascade framework exhibits high performance for early detection and localization of myocardial infarction, demonstrating potential as a clinical decision support tool.</div></div>","PeriodicalId":49836,"journal":{"name":"Medical Engineering & Physics","volume":"143 ","pages":"Article 104400"},"PeriodicalIF":2.3,"publicationDate":"2025-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144700158","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}

引用次数: 0