Biocybernetics and Biomedical Engineering最新文献_第2页

Effect of non-Newtonian rheological models on pulsatile hemodynamics in patients-specific venous models of pulsatile tinnitus 非牛顿流变学模型对搏动性耳鸣患者特异性静脉模型搏动血流动力学的影响

IF 6.6 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biocybernetics and Biomedical Engineering

Pub Date : 2025-10-01 DOI: 10.1016/j.bbe.2025.10.002

Zhenxia Mu , Ben Liu , Yicheng Han , Lihui Zhuang , Xiaoyu Qiu , Heyu Ding , Shusheng Gong , Guopeng Wang , Bin Gao , Youjun Liu , Shifeng Yang , Zhenchang Wang , Pengfei Zhao , Ximing Wang

Hemodynamic factors play a crucial role in the pathogenesis of venous pulsatile tinnitus (PT). The selection of an appropriate blood viscosity model is therefore essential for accurately capturing hemodynamic characteristics in numerical simulations. This study aimed to investigate and compare the effects of different blood rheology models on hemodynamic parameters in patients with venous PT. Numerical simulation of pulsatile blood flow was conducted in three-dimensional patient-specific models with sigmoid sinuses wall dehiscence (SSWD) accompanied by sigmoid sinuses diverticulum (SSD) or transverse sinus stenosis. Different blood rheology models were employed in the simulations, including the Newtonian, Power law, Carreau, and Herschel-Bulkley models. Results demonstrated that unfavorable hemodynamics, characterized by high-velocity patterns and abnormal distributions of wall pressure, wall shear stress (WSS), and time-average WSS (TAWSS) in specific SSD and SSWD regions, could increase the risk of venous PT. Both Newtonian and non-Newtonian models predicted comparable distributions of hemodynamic parameters. However, differences in magnitude were observed, particularly in the SSD and SSWD regions. The Power law model exhibited the most pronounced differences, predicting the lowest velocity in the SSD region and the highest wall pressure, WSS, and TAWSS in the SSWD region. The Herschel-Bulkley model showed similar trends but with less extreme magnitudes. The Carreau model was closely aligned with the Newtonian model. Although the Newtonian model generally predicted hemodynamic parameter distributions comparable to those of non-Newtonian models, marked differences were observed in key regions (SSD and SSWD) critically involved in venous PT pathogenesis. Therefore, selecting an appropriate viscosity model is essential for accurately assessing hemodynamic characteristics within these specific regions.

血流动力学因素在静脉搏动性耳鸣的发病机制中起着至关重要的作用。因此，选择合适的血液粘度模型对于在数值模拟中准确捕获血流动力学特性至关重要。本研究旨在探讨和比较不同血液流变学模型对静脉PT患者血流动力学参数的影响，在乙状窦壁裂（SSWD）伴乙状窦憩室（SSD）或横窦狭窄的三维患者特异性模型中，对脉动血流进行数值模拟。在模拟中采用了不同的血液流变学模型，包括牛顿模型、幂律模型、careau模型和Herschel-Bulkley模型。结果表明，在特定的SSD和SSWD区域，以高速模式和壁压、壁剪切应力（WSS）和时间平均WSS （TAWSS）的异常分布为特征的不利血流动力学可增加静脉PT的风险。牛顿模型和非牛顿模型预测的血流动力学参数分布相似。然而，观察到的幅度差异，特别是在SSD和SSWD区域。幂律模型表现出最明显的差异，预测了SSD区域的最低速度，以及SSWD区域的最高壁压、WSS和TAWSS。Herschel-Bulkley模型显示了类似的趋势，但幅度没有那么极端。卡罗模型与牛顿模型非常接近。尽管牛顿模型预测的血流动力学参数分布与非牛顿模型一般相当，但在静脉PT发病的关键区域（SSD和SSWD）中观察到显著差异。因此，选择合适的粘度模型对于准确评估这些特定区域的血流动力学特性至关重要。

{"title":"Effect of non-Newtonian rheological models on pulsatile hemodynamics in patients-specific venous models of pulsatile tinnitus","authors":"Zhenxia Mu , Ben Liu , Yicheng Han , Lihui Zhuang , Xiaoyu Qiu , Heyu Ding , Shusheng Gong , Guopeng Wang , Bin Gao , Youjun Liu , Shifeng Yang , Zhenchang Wang , Pengfei Zhao , Ximing Wang","doi":"10.1016/j.bbe.2025.10.002","DOIUrl":"10.1016/j.bbe.2025.10.002","url":null,"abstract":"<div><div>Hemodynamic factors play a crucial role in the pathogenesis of venous pulsatile tinnitus (PT). The selection of an appropriate blood viscosity model is therefore essential for accurately capturing hemodynamic characteristics in numerical simulations. This study aimed to investigate and compare the effects of different blood rheology models on hemodynamic parameters in patients with venous PT. Numerical simulation of pulsatile blood flow was conducted in three-dimensional patient-specific models with sigmoid sinuses wall dehiscence (SSWD) accompanied by sigmoid sinuses diverticulum (SSD) or transverse sinus stenosis. Different blood rheology models were employed in the simulations, including the Newtonian, Power law, Carreau, and Herschel-Bulkley models. Results demonstrated that unfavorable hemodynamics, characterized by high-velocity patterns and abnormal distributions of wall pressure, wall shear stress (WSS), and time-average WSS (TAWSS) in specific SSD and SSWD regions, could increase the risk of venous PT. Both Newtonian and non-Newtonian models predicted comparable distributions of hemodynamic parameters. However, differences in magnitude were observed, particularly in the SSD and SSWD regions. The Power law model exhibited the most pronounced differences, predicting the lowest velocity in the SSD region and the highest wall pressure, WSS, and TAWSS in the SSWD region. The Herschel-Bulkley model showed similar trends but with less extreme magnitudes. The Carreau model was closely aligned with the Newtonian model. Although the Newtonian model generally predicted hemodynamic parameter distributions comparable to those of non-Newtonian models, marked differences were observed in key regions (SSD and SSWD) critically involved in venous PT pathogenesis. Therefore, selecting an appropriate viscosity model is essential for accurately assessing hemodynamic characteristics within these specific regions.</div></div>","PeriodicalId":55381,"journal":{"name":"Biocybernetics and Biomedical Engineering","volume":"45 4","pages":"Pages 630-641"},"PeriodicalIF":6.6,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145362280","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}

引用次数: 0

Interferometric speckle contrast optical spectroscopy (iSCOS) in continuous-wave parallel interferometric near-infrared spectroscopy (CW-πNIRS) 连续波平行干涉近红外光谱（CW-πNIRS）中的干涉散斑对比光谱（iSCOS）

IF 6.6 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biocybernetics and Biomedical Engineering

Pub Date : 2025-10-01 DOI: 10.1016/j.bbe.2025.11.001

Klaudia Nowacka-Pieszak , Saeed Samaei , Dawid Borycki

Optical methods enable continuous, noninvasive cerebral blood flow (CBF) monitoring. Diffuse correlation spectroscopy (DCS) estimates CBF through temporal correlation analysis of scattered light but is limited by low detection throughput. Parallelizing DCS enhances performance but requires costly ultra-fast (∼1 MHz) detectors, complicating continuous measurements. An alternative approach analyzes spatial correlations using speckle contrast, inversely proportional to blood flow, captured with slower two-dimensional sensors. In this study, we present continuous-wave parallel interferometric near-infrared spectroscopy (CW-πNIRS), employing interferometry combined with a high-speed 2D camera, as a novel method uniquely suited for spatial correlation measurements. By leveraging interferometric detection, our approach provides a synthetic multi-exposure capability for direct quantitative comparisons between spatial (speckle contrast) and temporal (autocorrelation) methods for CBF monitoring. Numerical simulations, incorporating interferometric reference fields, and tissue-mimicking phantom validations demonstrated robust, and stable speckle contrast estimates. Finally, in vivo experiments confirmed the method’s potential for effective human cerebral blood flow monitoring, highlighting practical advantages and providing a clear pathway towards clinical implementation.

光学方法可以实现连续的、无创的脑血流（CBF）监测。漫射相关光谱学（DCS）通过散射光的时间相关分析来估计CBF，但受低检测吞吐量的限制。并行DCS提高了性能，但需要昂贵的超高速（~ 1 MHz）检测器，使连续测量复杂化。另一种方法是使用散斑对比度分析空间相关性，与血流成反比，用较慢的二维传感器捕获。在这项研究中，我们提出了连续波平行干涉近红外光谱（CW-πNIRS），该方法采用干涉测量与高速二维相机相结合，是一种独特适用于空间相关测量的新方法。通过利用干涉检测，我们的方法为CBF监测的空间（散斑对比）和时间（自相关）方法之间的直接定量比较提供了综合的多曝光能力。结合干涉参考场的数值模拟和组织模拟幻象验证证明了鲁棒和稳定的散斑对比度估计。最后，体内实验证实了该方法有效监测人脑血流量的潜力，突出了实用性优势，为临床应用提供了明确的途径。

{"title":"Interferometric speckle contrast optical spectroscopy (iSCOS) in continuous-wave parallel interferometric near-infrared spectroscopy (CW-πNIRS)","authors":"Klaudia Nowacka-Pieszak , Saeed Samaei , Dawid Borycki","doi":"10.1016/j.bbe.2025.11.001","DOIUrl":"10.1016/j.bbe.2025.11.001","url":null,"abstract":"<div><div>Optical methods enable continuous, noninvasive cerebral blood flow (CBF) monitoring. Diffuse correlation spectroscopy (DCS) estimates CBF through temporal correlation analysis of scattered light but is limited by low detection throughput. Parallelizing DCS enhances performance but requires costly ultra-fast (∼1 MHz) detectors, complicating continuous measurements. An alternative approach analyzes spatial correlations using speckle contrast, inversely proportional to blood flow, captured with slower two-dimensional sensors. In this study, we present continuous-wave parallel interferometric near-infrared spectroscopy (CW-πNIRS), employing interferometry combined with a high-speed 2D camera, as a novel method uniquely suited for spatial correlation measurements. By leveraging interferometric detection, our approach provides a synthetic multi-exposure capability for direct quantitative comparisons between spatial (speckle contrast) and temporal (autocorrelation) methods for CBF monitoring. Numerical simulations, incorporating interferometric reference fields, and tissue-mimicking phantom validations demonstrated robust, and stable speckle contrast estimates. Finally, <em>in vivo</em> experiments confirmed the method’s potential for effective human cerebral blood flow monitoring, highlighting practical advantages and providing a clear pathway towards clinical implementation.</div></div>","PeriodicalId":55381,"journal":{"name":"Biocybernetics and Biomedical Engineering","volume":"45 4","pages":"Pages 669-684"},"PeriodicalIF":6.6,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145465358","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}

引用次数: 0

Enhanced computer-aided system for breast lesion classification and grading using novel radio frequency time series approach 使用新型射频时间序列方法增强乳腺病变分类和分级的计算机辅助系统

IF 6.6 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biocybernetics and Biomedical Engineering

Pub Date : 2025-10-01 DOI: 10.1016/j.bbe.2025.10.003

Elaheh Norouzi Ghehi , Ali Fallah , Saeid Rashidi , Maryam Mehdizadeh Dastjerdi

Objective

Breast cancer is the most common malignancy among women and leading cause of mortality. Accurate, non-invasive differentiation of benign and malignant lesions is a clinical priority to reduce unnecessary biopsies and enable timely treatment. Elastography and RF time series (RF TS) processing are effective ultrasound-based techniques for tissue characterization. To improve their accuracy, we introduced an innovative approach called RFTSDP (RF Time Series Dynamic Processing). In RFTSDP, data are recorded during mechanical stimulation, revealing tissue properties in RF echoes. Extracting relevant features enhances computer-aided methods and improves tissue classification and grading.

Materials and methods

An implement was developed to induce vibrations at different frequencies. Data were collected from ex-vivo tissues embedded in normal mimicking phantoms. Raw focused, raw, and beamformed ultrafast data were recorded under no stimulation, constant force, and various vibrational stimulations using the Supersonic Imaging Aixplorer ultrasound system. Features were extracted from each RF TS across the time, time–frequency, spectral, and non-linear domains. Multiple classifiers were evaluated, among which support vector machines with different kernels achieved the best results.

Results

Beyond the classification of cancerous versus non-cancerous tissue, we also classified different cancerous lesion types and graded invasive ductal carcinoma. The best results were achieved with beamformed ultrafast data under 65 Hz vibrational stimulation. The mean classification accuracies for 2-, 3-, and 5-class were 99.78 %, 99.06 % and 99.32 %, respectively.

Conclusions

The outcomes affirm that applying vibration, particularly at an optimal frequency, enhances breast tissue classification. The proposed method demonstrated efficacy not only in distinguishing between cancerous and non-cancerous lesions but also in grading cancerous tissues.

目的乳腺癌是妇女中最常见的恶性肿瘤，也是导致死亡的主要原因。准确、无创地鉴别良恶性病变是减少不必要的活检和及时治疗的临床重点。弹性成像和射频时间序列（RF TS）处理是有效的基于超声的组织表征技术。为了提高它们的精度，我们引入了一种称为RFTSDP （RF时间序列动态处理）的创新方法。在RFTSDP中，在机械刺激过程中记录数据，揭示射频回波中的组织特性。提取相关特征增强了计算机辅助方法，改进了组织分类和分级。材料和方法研制了一种装置来诱导不同频率的振动。数据收集自植入正常模拟模型的离体组织。使用超声成像aiexplorer超声系统，在无刺激、恒力和各种振动刺激下记录原始聚焦、原始和波束形成的超快数据。从每个RF TS中提取时间、时频、频谱和非线性域的特征。对多个分类器进行了评价，其中具有不同核的支持向量机获得了最好的分类效果。结果除了癌组织与非癌组织的分类外，我们还对浸润性导管癌进行了不同类型和分级的分类。在65 Hz的振动刺激下，波束形成的超快数据获得了最好的结果。2类、3类和5类的平均分类准确率分别为99.78%、99.06%和99.32%。结论应用振动，特别是在最佳频率下，可以增强乳腺组织分类。所提出的方法不仅在区分癌性和非癌性病变方面有效，而且在癌性组织分级方面也有效。

{"title":"Enhanced computer-aided system for breast lesion classification and grading using novel radio frequency time series approach","authors":"Elaheh Norouzi Ghehi , Ali Fallah , Saeid Rashidi , Maryam Mehdizadeh Dastjerdi","doi":"10.1016/j.bbe.2025.10.003","DOIUrl":"10.1016/j.bbe.2025.10.003","url":null,"abstract":"<div><h3>Objective</h3><div>Breast cancer is the most common malignancy among women and leading cause of mortality. Accurate, non-invasive differentiation of benign and malignant lesions is a clinical priority to reduce unnecessary biopsies and enable timely treatment. Elastography and RF time series (RF TS) processing are effective ultrasound-based techniques for tissue characterization. To improve their accuracy, we introduced an innovative approach called RFTSDP (RF Time Series Dynamic Processing). In RFTSDP, data are recorded during mechanical stimulation, revealing tissue properties in RF echoes. Extracting relevant features enhances computer-aided methods and improves tissue classification and grading.</div></div><div><h3>Materials and methods</h3><div>An implement was developed to induce vibrations at different frequencies. Data were collected from ex-vivo tissues embedded in normal mimicking phantoms. Raw focused, raw, and beamformed ultrafast data were recorded under no stimulation, constant force, and various vibrational stimulations using the Supersonic Imaging Aixplorer ultrasound system. Features were extracted from each RF TS across the time, time–frequency, spectral, and non-linear domains. Multiple classifiers were evaluated, among which support vector machines with different kernels achieved the best results.</div></div><div><h3>Results</h3><div>Beyond the classification of cancerous versus non-cancerous tissue, we also classified different cancerous lesion types and graded invasive ductal carcinoma. The best results were achieved with beamformed ultrafast data under 65 Hz vibrational stimulation. The mean classification accuracies for 2-, 3-, and 5-class were 99.78 %, 99.06 % and 99.32 %, respectively.</div></div><div><h3>Conclusions</h3><div>The outcomes affirm that applying vibration, particularly at an optimal frequency, enhances breast tissue classification. The proposed method demonstrated efficacy not only in distinguishing between cancerous and non-cancerous lesions but also in grading cancerous tissues.</div></div>","PeriodicalId":55381,"journal":{"name":"Biocybernetics and Biomedical Engineering","volume":"45 4","pages":"Pages 655-668"},"PeriodicalIF":6.6,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145416928","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}

引用次数: 0

The influence of tissue conductivity uncertainty on the nerve activation thresholds in non-invasive electrical phrenic nerve stimulation 组织电导率不确定性对无创膈神经电刺激神经激活阈值的影响

IF 6.6 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biocybernetics and Biomedical Engineering

Pub Date : 2025-10-01 DOI: 10.1016/j.bbe.2025.11.003

Laureen Wegert , Luca Di Rienzo , Lorenzo Codecasa , Sicheng An , Marek Ziolkowski , Alexander Hunold , Irene Lange , Tim Kalla , Jens Haueisen

Non-invasive phrenic nerve stimulation can be used to overcome diaphragm insufficiency caused by mechanical ventilation. Detailed models and electromagnetic simulations are used to suggest appropriate stimulation parameters, but require accurate tissue properties. However, a wide range of electrical conductivity values is known from the literature. Here, we aim to perform an uncertainty analysis of the nerve activation threshold and the potential distribution along the phrenic nerve due to uncertain tissue conductivites.

We built a generalized polynomial chaos (gPC) model to calculate the phrenic nerve activation threshold. It was based on a reduced order model of a detailed anatomical finite element model of the neck including 13 tissue types to calculate the potential distribution, followed by a biophysiological nerve model. The tissue conductivity values investigated here were for the compartments of fat, muscle, nerve, and soft tissue. Their influence on the nerve activation threshold was investigated by changing conductivity values of the single tissues and all tissues at a time within a Monte Carlo analysis using the gPC model.

The phrenic nerve activation threshold varied between 33.8 mA and 46.9 mA for the combined variation of the conductivity values. Sobol indices and global sensitivity coefficients indicated the highest influence for muscle conductivity value, followed by soft tissue, fat, and nerve tissue.

Our results may have implications for understanding the experimentally observed variation in individual phrenic nerve activation thresholds affected by physiological and pathological conductivity changes. Accurate electric properties of muscle and soft tissue and detailed geometric representations should be considered in electromagnetic simulations.

无创膈神经刺激可用于克服机械通气引起的膈肌功能不全。详细的模型和电磁模拟用于建议适当的刺激参数，但需要准确的组织特性。然而，从文献中已知电导率值的范围很广。在这里，我们的目的是进行神经激活阈值的不确定性分析和电位分布沿膈神经由于不确定的组织电导率。我们建立了广义多项式混沌（gPC）模型来计算膈神经的激活阈值。该方法基于包含13种组织类型的颈部详细解剖有限元模型的降阶模型计算电位分布，然后建立生物生理神经模型。这里研究的组织电导率值是针对脂肪、肌肉、神经和软组织的隔室。在使用gPC模型的蒙特卡罗分析中，通过改变单个组织和所有组织的电导率值，研究了它们对神经激活阈值的影响。电导率综合变化的膈神经激活阈值在33.8 ~ 46.9 mA之间。Sobol指数和整体敏感性系数对肌肉电导率的影响最大，其次是软组织、脂肪和神经组织。我们的结果可能有助于理解实验观察到的受生理和病理电导率变化影响的个体膈神经激活阈值的变化。电磁仿真应考虑肌肉和软组织的准确电特性和详细的几何表示。

{"title":"The influence of tissue conductivity uncertainty on the nerve activation thresholds in non-invasive electrical phrenic nerve stimulation","authors":"Laureen Wegert , Luca Di Rienzo , Lorenzo Codecasa , Sicheng An , Marek Ziolkowski , Alexander Hunold , Irene Lange , Tim Kalla , Jens Haueisen","doi":"10.1016/j.bbe.2025.11.003","DOIUrl":"10.1016/j.bbe.2025.11.003","url":null,"abstract":"<div><div>Non-invasive phrenic nerve stimulation can be used to overcome diaphragm insufficiency caused by mechanical ventilation. Detailed models and electromagnetic simulations are used to suggest appropriate stimulation parameters, but require accurate tissue properties. However, a wide range of electrical conductivity values is known from the literature. Here, we aim to perform an uncertainty analysis of the nerve activation threshold and the potential distribution along the phrenic nerve due to uncertain tissue conductivites.</div><div>We built a generalized polynomial chaos (gPC) model to calculate the phrenic nerve activation threshold. It was based on a reduced order model of a detailed anatomical finite element model of the neck including 13 tissue types to calculate the potential distribution, followed by a biophysiological nerve model. The tissue conductivity values investigated here were for the compartments of fat, muscle, nerve, and soft tissue. Their influence on the nerve activation threshold was investigated by changing conductivity values of the single tissues and all tissues at a time within a Monte Carlo analysis using the gPC model.</div><div>The phrenic nerve activation threshold varied between 33.8 mA and 46.9 mA for the combined variation of the conductivity values. Sobol indices and global sensitivity coefficients indicated the highest influence for muscle conductivity value, followed by soft tissue, fat, and nerve tissue.</div><div>Our results may have implications for understanding the experimentally observed variation in individual phrenic nerve activation thresholds affected by physiological and pathological conductivity changes. Accurate electric properties of muscle and soft tissue and detailed geometric representations should be considered in electromagnetic simulations.</div></div>","PeriodicalId":55381,"journal":{"name":"Biocybernetics and Biomedical Engineering","volume":"45 4","pages":"Pages 697-706"},"PeriodicalIF":6.6,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145617673","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}

引用次数: 0

Estimating core body temperature from heart rate using a residual-compensated adaptive Kalman filter 利用残差补偿自适应卡尔曼滤波从心率估计核心体温

IF 6.6 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biocybernetics and Biomedical Engineering

Pub Date : 2025-10-01 DOI: 10.1016/j.bbe.2025.10.001

Yuanzhe Zhao , Jeroen H.M. Bergmann

Accurate, real-time estimation of core body temperature is critical for preventing heat-related illness. While existing Kalman filter-based methods offer interpretable, single-input (heart rate) solutions, they are limited by fixed observation models that fail to capture the complex, non-linear, state-dependent dynamics of physiological signals.

To address this, we propose the Residual-Compensated Adaptive Kalman Filter (RCAKF), a novel hybrid framework. The RCAKF integrates a long short-term memory (LSTM) network to learn and correct structured, state-dependent errors in the observation model, alongside an adaptive noise estimator that dynamically adjusts for measurement uncertainty. This architecture enhances the classic Kalman filter with data-driven flexibility while maintaining its recursive structure and interpretability.

Evaluation was conducted on a controlled experimental dataset with 22 participants performing exercise and recovery under varied thermal conditions. Compared to five baseline models: extended Kalman filter (EKF: RMSE = 0.39 °C), the improved ECTemp model with a sigmoid observation function (ECTemp-S: RMSE = 0.40 °C), biphasic Kalman filter-based model (BKFB: RMSE = 0.48 °C), moving-average Kalman filter (MAKF: RMSE = 0.38 °C), and a standalone LSTM network (RMSE = 0.46 °C), RCAKF achieved the best accuracy with an RMSE of 0.31 °C.

By augmenting the Kalman filter with a learned residual correction and adaptive uncertainty, the RCAKF framework significantly enhances core temperature tracking from a single heart rate signal. Its accuracy and reliance on a single, common sensor make it a practical and promising solution for real-time deployment on wearable devices for safety monitoring.

准确、实时地估计核心体温对于预防与热有关的疾病至关重要。虽然现有的基于卡尔曼滤波的方法提供了可解释的单输入（心率）解决方案，但它们受到固定观察模型的限制，这些模型无法捕捉生理信号的复杂、非线性、状态依赖的动态。为了解决这个问题，我们提出了一种新的混合框架残余补偿自适应卡尔曼滤波器（RCAKF）。RCAKF集成了一个长短期记忆（LSTM）网络，用于学习和纠正观测模型中的结构化、状态相关错误，以及一个自适应噪声估计器，可动态调整测量不确定性。该结构增强了经典卡尔曼滤波器的数据驱动灵活性，同时保持了其递归结构和可解释性。对22名参与者在不同热条件下进行运动和恢复的受控实验数据集进行了评估。与扩展卡尔曼滤波（EKF: RMSE = 0.39°C）、具有s型观测函数的改进ECTemp模型（ECTemp- s: RMSE = 0.40°C）、基于双相卡尔曼滤波的模型（BKFB: RMSE = 0.48°C）、移动平均卡尔曼滤波（MAKF: RMSE = 0.38°C）和独立LSTM网络（RMSE = 0.46°C）等5种基线模型相比，RCAKF的RMSE为0.31°C，达到了最佳精度。通过使用学习残差校正和自适应不确定性增强卡尔曼滤波器，RCAKF框架显著增强了单次心率信号的核心温度跟踪。其准确性和对单个通用传感器的依赖使其成为可穿戴设备实时部署的实用且有前途的安全监控解决方案。

{"title":"Estimating core body temperature from heart rate using a residual-compensated adaptive Kalman filter","authors":"Yuanzhe Zhao , Jeroen H.M. Bergmann","doi":"10.1016/j.bbe.2025.10.001","DOIUrl":"10.1016/j.bbe.2025.10.001","url":null,"abstract":"<div><div>Accurate, real-time estimation of core body temperature is critical for preventing heat-related illness. While existing Kalman filter-based methods offer interpretable, single-input (heart rate) solutions, they are limited by fixed observation models that fail to capture the complex, non-linear, state-dependent dynamics of physiological signals.</div><div>To address this, we propose the Residual-Compensated Adaptive Kalman Filter (RCAKF), a novel hybrid framework. The RCAKF integrates a long short-term memory (LSTM) network to learn and correct structured, state-dependent errors in the observation model, alongside an adaptive noise estimator that dynamically adjusts for measurement uncertainty. This architecture enhances the classic Kalman filter with data-driven flexibility while maintaining its recursive structure and interpretability.</div><div>Evaluation was conducted on a controlled experimental dataset with 22 participants performing exercise and recovery under varied thermal conditions. Compared to five baseline models: extended Kalman filter (EKF: RMSE = 0.39 °C), the improved ECTemp model with a sigmoid observation function (ECTemp-S: RMSE = 0.40 °C), biphasic Kalman filter-based model (BKFB: RMSE = 0.48 °C), moving-average Kalman filter (MAKF: RMSE = 0.38 °C), and a standalone LSTM network (RMSE = 0.46 °C), RCAKF achieved the best accuracy with an RMSE of 0.31 °C.</div><div>By augmenting the Kalman filter with a learned residual correction and adaptive uncertainty, the RCAKF framework significantly enhances core temperature tracking from a single heart rate signal. Its accuracy and reliance on a single, common sensor make it a practical and promising solution for real-time deployment on wearable devices for safety monitoring.</div></div>","PeriodicalId":55381,"journal":{"name":"Biocybernetics and Biomedical Engineering","volume":"45 4","pages":"Pages 617-629"},"PeriodicalIF":6.6,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145323893","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}

引用次数: 0

Reevaluating performance in c-VEP BCIs: The impact of calibration time 重新评估c-VEP bci的性能：校准时间的影响

IF 6.6 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biocybernetics and Biomedical Engineering

Pub Date : 2025-10-01 DOI: 10.1016/j.bbe.2025.10.006

Víctor Martínez-Cagigal , Eduardo Santamaría-Vázquez , Sergio Pérez-Velasco , Ana Martín-Fernández , Roberto Hornero

Code-modulated visual evoked potentials (c-VEP) have demonstrated high performance in non-invasive brain-computer interfaces (BCIs). Recently, research has begun to consider practical aspects such as visual comfort, where non-binary sequences and variations in the spatial frequency of stimuli play significant roles. However, calibration requirements remain underexplored in performance comparisons. This study aims to analyze a multi-variable tradeoff crucial to the practical application of c-VEP-based BCIs: decoding accuracy, decoding speed, and calibration time. Visual comfort is retrospectively evaluated using two pre-recorded datasets. Models were trained with increasing calibration cycles and tested across varying decoding times, depicting learning and decoding curves. The datasets comprised 32 healthy subjects, and featured different stimulus paradigms: plain non-binary stimuli and checkerboard-like binary stimuli with spatial frequency variations. Results showed that all conditions achieved over 97 % grand-averaged accuracy with sufficient calibration. However, a clear tradeoff emerged between calibration duration and performance. Achieving 95 % average accuracy within a 2 s decoding window required mean calibration durations of 28.7 ± 19.0 s for binary stimuli, or 148.7 ± 72.3 s for non-binary stimuli. The binary checkerboard-based condition with a spatial frequency of 1.2 c/º (C016) proved to be particularly effective, achieving over 95 % accuracy within 2 s decoding window using only 7.3 s of calibration, and reporting a significant improvement in visual comfort. A minimum calibration time of 1 min was considered essential to adequately estimate the brain response, critical in template-matching paradigms. In conclusion, achieving optimal c-VEP performance requires balancing calibration duration, decoding speed and accuracy, and visual comfort.

编码调制视觉诱发电位（c-VEP）在非侵入性脑机接口（bci）中表现出很高的性能。最近，研究开始考虑视觉舒适等实际方面，其中非二进制序列和刺激空间频率的变化起着重要作用。然而，在性能比较中，校准要求仍未得到充分探讨。本研究旨在分析对基于c- vep的bci的实际应用至关重要的多变量权衡：解码精度，解码速度和校准时间。使用两个预先记录的数据集对视觉舒适度进行回顾性评估。模型通过不断增加的校准周期进行训练，并在不同的解码时间内进行测试，描绘学习和解码曲线。数据集包括32名健康受试者，具有不同的刺激范式：普通的非二进制刺激和具有空间频率变化的棋盘状二进制刺激。结果表明，在充分校准的情况下，所有条件的大平均精度均达到97%以上。然而，在校准时间和性能之间出现了明显的权衡。要在2秒的解码窗口内达到95%的平均精度，二进制刺激的平均校准时间为28.7 ± 19.0 s，非二进制刺激的平均校准时间为148.7 ± 72.3 s。基于二进制棋盘的空间频率为1.2 c/º（C016）的条件被证明是特别有效的，仅使用7.3秒的校准时间，在2秒的解码窗口内达到95%以上的准确率，并报告了视觉舒适度的显着改善。最小校准时间1 min被认为是充分估计大脑反应的必要条件，在模板匹配范式中至关重要。总之，实现最佳的c-VEP性能需要平衡校准时间、解码速度和精度以及视觉舒适性。

{"title":"Reevaluating performance in c-VEP BCIs: The impact of calibration time","authors":"Víctor Martínez-Cagigal , Eduardo Santamaría-Vázquez , Sergio Pérez-Velasco , Ana Martín-Fernández , Roberto Hornero","doi":"10.1016/j.bbe.2025.10.006","DOIUrl":"10.1016/j.bbe.2025.10.006","url":null,"abstract":"<div><div>Code-modulated visual evoked potentials (c-VEP) have demonstrated high performance in non-invasive brain-computer interfaces (BCIs). Recently, research has begun to consider practical aspects such as visual comfort, where non-binary sequences and variations in the spatial frequency of stimuli play significant roles. However, calibration requirements remain underexplored in performance comparisons. This study aims to analyze a multi-variable tradeoff crucial to the practical application of c-VEP-based BCIs: decoding accuracy, decoding speed, and calibration time. Visual comfort is retrospectively evaluated using two pre-recorded datasets. Models were trained with increasing calibration cycles and tested across varying decoding times, depicting learning and decoding curves. The datasets comprised 32 healthy subjects, and featured different stimulus paradigms: plain non-binary stimuli and checkerboard-like binary stimuli with spatial frequency variations. Results showed that all conditions achieved over 97 % grand-averaged accuracy with sufficient calibration. However, a clear tradeoff emerged between calibration duration and performance. Achieving 95 % average accuracy within a 2 s decoding window required mean calibration durations of 28.7 ± 19.0 s for binary stimuli, or 148.7 ± 72.3 s for non-binary stimuli. The binary checkerboard-based condition with a spatial frequency of 1.2 c/º (C016) proved to be particularly effective, achieving over 95 % accuracy within 2 s decoding window using only 7.3 s of calibration, and reporting a significant improvement in visual comfort. A minimum calibration time of 1 min was considered essential to adequately estimate the brain response, critical in template-matching paradigms. In conclusion, achieving optimal c-VEP performance requires balancing calibration duration, decoding speed and accuracy, and visual comfort.</div></div>","PeriodicalId":55381,"journal":{"name":"Biocybernetics and Biomedical Engineering","volume":"45 4","pages":"Pages 685-696"},"PeriodicalIF":6.6,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145519677","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}

引用次数: 0

A novel class of rank tests for high-dimensional data with an application to Alzheimer’s disease 一类新的高维数据秩检验及其在阿尔茨海默病中的应用

IF 6.6 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biocybernetics and Biomedical Engineering

Pub Date : 2025-10-01 DOI: 10.1016/j.bbe.2025.10.005

Jan Kalina , Jaromir Kukal , Oldrich Vysata

Electroencephalography (EEG) signals are widely used in neuroscience research and biomedical applications, particularly for comparative analyses between two cohorts: patients and control subjects. In EEG research, multivariate two-sample tests remain underused, while multiple comparison procedures are often misapplied as a substitute, leading to inadequate error control. Moreover, existing rank-based tests rely solely on Euclidean distances and lack robustness to outliers. The objective of this work is to develop and evaluate a new class of rank-based tests for high-dimensional data that employ robust Mahalanobis interpoint distances, and to demonstrate their practical value in the analysis of EEG signals. This class includes a version based on the MRWCD (minimum regularized weighted covariance determinant) estimator, which enhances the robustness of the Mahalanobis distances by mitigating the influence of outliers.

To illustrate the effectiveness of these tests, EEG data consisting of 1216 variables from 28 patients with Alzheimer’s disease and 146 healthy control individuals were analyzed. The results of multivariate tests reveal significant findings, which are also explored in the context of individual EEG channels and frequencies. Among the approaches tested, rank-based tests using the newly proposed interpoint distances, particularly in combination with the Cucconi test statistic, yield the strongest results. For the comparison between patients and controls, a p-value of 0.004 was obtained, which is below the significance level

α = 0.05

, indicating a statistically significant difference.

脑电图（EEG）信号广泛应用于神经科学研究和生物医学应用，特别是用于患者和对照受试者两个队列之间的比较分析。在脑电图研究中，多变量双样本测试仍未得到充分利用，而多种比较程序往往被误用作为替代，导致误差控制不足。此外，现有的基于秩的测试仅依赖于欧几里得距离，对异常值缺乏鲁棒性。这项工作的目的是开发和评估一类新的基于秩的高维数据测试，该测试采用稳健的马氏点间距离，并展示其在脑电图信号分析中的实用价值。该类包括一个基于MRWCD（最小正则化加权协方差行列式）估计器的版本，它通过减轻异常值的影响来增强马氏距离的鲁棒性。为了说明这些测试的有效性，我们分析了来自28名阿尔茨海默病患者和146名健康对照者的1216个变量的脑电图数据。多变量测试的结果揭示了显著的发现，这些发现也在个体脑电图通道和频率的背景下进行了探讨。在测试的方法中，使用新提出的点间距离的基于等级的测试，特别是与Cucconi测试统计量相结合，产生了最强的结果。患者与对照组比较，p值为0.004，低于显著性水平α=0.05，差异有统计学意义。

{"title":"A novel class of rank tests for high-dimensional data with an application to Alzheimer’s disease","authors":"Jan Kalina , Jaromir Kukal , Oldrich Vysata","doi":"10.1016/j.bbe.2025.10.005","DOIUrl":"10.1016/j.bbe.2025.10.005","url":null,"abstract":"<div><div>Electroencephalography (EEG) signals are widely used in neuroscience research and biomedical applications, particularly for comparative analyses between two cohorts: patients and control subjects. In EEG research, multivariate two-sample tests remain underused, while multiple comparison procedures are often misapplied as a substitute, leading to inadequate error control. Moreover, existing rank-based tests rely solely on Euclidean distances and lack robustness to outliers. The objective of this work is to develop and evaluate a new class of rank-based tests for high-dimensional data that employ robust Mahalanobis interpoint distances, and to demonstrate their practical value in the analysis of EEG signals. This class includes a version based on the MRWCD (minimum regularized weighted covariance determinant) estimator, which enhances the robustness of the Mahalanobis distances by mitigating the influence of outliers.</div><div>To illustrate the effectiveness of these tests, EEG data consisting of 1216 variables from 28 patients with Alzheimer’s disease and 146 healthy control individuals were analyzed. The results of multivariate tests reveal significant findings, which are also explored in the context of individual EEG channels and frequencies. Among the approaches tested, rank-based tests using the newly proposed interpoint distances, particularly in combination with the Cucconi test statistic, yield the strongest results. For the comparison between patients and controls, a <em>p</em>-value of 0.004 was obtained, which is below the significance level <span><math><mrow><mi>α</mi><mo>=</mo><mn>0.05</mn></mrow></math></span>, indicating a statistically significant difference.</div></div>","PeriodicalId":55381,"journal":{"name":"Biocybernetics and Biomedical Engineering","volume":"45 4","pages":"Pages 707-717"},"PeriodicalIF":6.6,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145617670","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}

引用次数: 0

A morphometric lung model for evaluating lung aeration at birth 一种用于评估出生时肺通气的肺形态计量学模型

IF 6.6 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biocybernetics and Biomedical Engineering

Pub Date : 2025-10-01 DOI: 10.1016/j.bbe.2025.11.004

Chiara Veneroni , Enrico Conca , Davide Bizzotto , Kenneth Lutchen , Alberto Tosi , Raffaele L. Dellacà

Background

Lung aeration must rapidly develop at birth, but how to promote it without injuring the fragile lung is still unknown. Mathematical models simulating lung mechanics during this transition may help understand the underlying physiological mechanisms and design protective ventilation strategies. This study develops a morphologically-coherent computational lung model incorporating changing physical conditions during the transition from liquid-filled to gas-filled lungs.

Methods

We adapted a 3-D morphological model of the adult airway tree adjusting airway dimensions, lung volume, and lung tissue mechanical properties. Changes in resistance, inertia, and compliance during aeration were modeled by considering the differing properties of fetal fluid versus air. The capillary pressure at the liquid–air interface was computed using Laplace equation. Terminal airway diameters increased with lung volume due to airway-parenchymal interdependence. An integrated circuit simulator solved the entire network in the time domain.

Results

The air volume entering the model at different applied pressures increased exponentially with time. With 30 cmH₂O applied, lung volume reached total capacity after 15 s, matching lung aeration dynamics observed in animal models and human infants. In contrast, after the same time, at 15 cmH₂O, lung volume was slightly above functional residual capacity, and at 10 cmH₂O, it remained below.

Conclusions

The proposed in-silico newborn lung model simulates lung aeration at birth, allowing observation of the airway emptying sequence and the heterogeneity of aeration at each time point. Integrating this model with comprehensive acinar models may aid in defining protective resuscitation ventilation strategies for recruiting the lung minimizing risk of injuries.

出生时肺通气必须迅速发展，但如何在不伤害脆弱的肺的情况下促进肺通气仍是未知的。在这一转变过程中模拟肺力学的数学模型可能有助于理解潜在的生理机制和设计保护性通气策略。本研究开发了一个形态学一致的计算肺模型，该模型包含了从充液肺到充气肺过渡过程中不断变化的物理条件。方法采用成人气道树三维形态模型，调整气道尺寸、肺体积和肺组织力学特性。通过考虑胎液与空气的不同特性，模拟了通气过程中阻力、惯性和顺应性的变化。利用拉普拉斯方程计算了气液界面处的毛细压力。末端气道直径随着肺容量的增加而增加，这是由于气道与肺实质的相互依赖关系。一个集成电路模拟器在时域内求解了整个网络。结果在不同施加压力下进入模型的风量随时间呈指数增长。应用30 cmH2O时，肺容量在15 s后达到总容量，与动物模型和人类婴儿观察到的肺通气动态相匹配。相比之下，相同时间后，在15 cmH2O时，肺容量略高于功能剩余容量，而在10 cmH2O时，肺容量仍低于功能剩余容量。结论所建立的计算机新生儿肺模型模拟了新生儿出生时的肺通气，可以观察气道排空顺序和各时间点通气的异质性。将该模型与全面的腺泡模型相结合，可能有助于确定保护性复苏通气策略，以最大限度地降低肺损伤风险。

{"title":"A morphometric lung model for evaluating lung aeration at birth","authors":"Chiara Veneroni , Enrico Conca , Davide Bizzotto , Kenneth Lutchen , Alberto Tosi , Raffaele L. Dellacà","doi":"10.1016/j.bbe.2025.11.004","DOIUrl":"10.1016/j.bbe.2025.11.004","url":null,"abstract":"<div><h3>Background</h3><div>Lung aeration must rapidly develop at birth, but how to promote it without injuring the fragile lung is still unknown. Mathematical models simulating lung mechanics during this transition may help understand the underlying physiological mechanisms and design protective ventilation strategies. This study develops a morphologically-coherent computational lung model incorporating changing physical conditions during the transition from liquid-filled to gas-filled lungs.</div></div><div><h3>Methods</h3><div>We adapted a 3-D morphological model of the adult airway tree adjusting airway dimensions, lung volume, and lung tissue mechanical properties. Changes in resistance, inertia, and compliance during aeration were modeled by considering the differing properties of fetal fluid versus air. The capillary pressure at the liquid–air interface was computed using Laplace equation. Terminal airway diameters increased with lung volume due to airway-parenchymal interdependence. An integrated circuit simulator solved the entire network in the time domain.</div></div><div><h3>Results</h3><div>The air volume entering the model at different applied pressures increased exponentially with time. With 30 cmH<sub>2</sub>O applied, lung volume reached total capacity after 15 s, matching lung aeration dynamics observed in animal models and human infants. In contrast, after the same time, at 15 cmH<sub>2</sub>O, lung volume was slightly above functional residual capacity, and at 10 cmH<sub>2</sub>O, it remained below.</div></div><div><h3>Conclusions</h3><div>The proposed in-silico newborn lung model simulates lung aeration at birth, allowing observation of the airway emptying sequence and the heterogeneity of aeration at each time point. Integrating this model with comprehensive acinar models may aid in defining protective resuscitation ventilation strategies for recruiting the lung minimizing risk of injuries.</div></div>","PeriodicalId":55381,"journal":{"name":"Biocybernetics and Biomedical Engineering","volume":"45 4","pages":"Pages 718-725"},"PeriodicalIF":6.6,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145617672","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}

引用次数: 0

Dual-peak cyclic loading for evaluation of lower-limb prostheses 双峰循环载荷对下肢假体的评价

IF 6.6 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biocybernetics and Biomedical Engineering

Pub Date : 2025-10-01 DOI: 10.1016/j.bbe.2025.10.004

Zhaokun Zhang , Wangrui Xu , Hanyu Gan , Chunlei K. Song , Albert J. Shih

Lower-limb prosthesis (LLP) is critical for the mobility and quality of life of amputees. Ensuring the reliability and durability of LLPs is essential to users’ safety, comfort, and mobility. While the current standard (ISO 10328) for the mechanical test of LLPs has served as a foundation, the sinusoidal cyclic loading method used for fatigue testing does not replicate the actual loading conditions on LLPs during a human walking gait. In normal walking, LLPs are subjected to a dual-peak loading condition with two distinct force peaks at the heel-strike and toe-off phases. Such a cyclic and dual-peak dynamic loading pattern on LLP is essential to test the durability of the LLP effectively and reliably. In this study, two simple and effective dual-peak cyclic loading test apparatuses are designed and built to evaluate the durability and functionality of LLPs. Two loading plates are used to contact the heel and toe of the prosthetic foot to simulate the dual-peak heel-strike and toe-off loadings between the foot and the ground. This dual-peak loading on the LLP is controlled and replicated by adjusting the positions of two loading plates and the actuator stroke to change the contact forces. Experimental results show that the proposed testing apparatuses and procedures can emulate the dual-peak axial loading of the LLP during normal walking gait, providing a more accurate testing method of the dynamic loading condition on LLPs than the current ISO standard.

下肢假肢（LLP）对截肢者的活动能力和生活质量至关重要。确保llp的可靠性和耐用性对用户的安全性、舒适性和移动性至关重要。虽然目前的llp力学测试标准（ISO 10328）已经作为基础，但用于疲劳测试的正弦循环加载方法并不能复制人类行走步态时llp的实际加载条件。在正常行走中，llp承受双峰加载条件，在脚跟撞击和脚趾撞击阶段有两个不同的力峰值。这种循环双峰动加载模式对于有效、可靠地测试LLP的耐久性至关重要。本研究设计并搭建了两台简单有效的双峰循环加载试验装置，以评估llp的耐久性和功能性。用两个加载板连接假肢脚的脚跟和脚趾，模拟脚与地面之间的双峰脚跟撞击和脚趾脱落载荷。通过调整两个加载板的位置和执行器行程来改变接触力，可以控制和复制LLP上的双峰载荷。实验结果表明，所提出的测试设备和程序能够模拟LLP在正常行走步态下的双峰轴向载荷，为LLP动态加载状况提供了比现行ISO标准更准确的测试方法。

{"title":"Dual-peak cyclic loading for evaluation of lower-limb prostheses","authors":"Zhaokun Zhang , Wangrui Xu , Hanyu Gan , Chunlei K. Song , Albert J. Shih","doi":"10.1016/j.bbe.2025.10.004","DOIUrl":"10.1016/j.bbe.2025.10.004","url":null,"abstract":"<div><div>Lower-limb prosthesis (LLP) is critical for the mobility and quality of life of amputees. Ensuring the reliability and durability of LLPs is essential to users’ safety, comfort, and mobility. While the current standard (ISO 10328) for the mechanical test of LLPs has served as a foundation, the sinusoidal cyclic loading method used for fatigue testing does not replicate the actual loading conditions on LLPs during a human walking gait. In normal walking, LLPs are subjected to a dual-peak loading condition with two distinct force peaks at the heel-strike and toe-off phases. Such a cyclic and dual-peak dynamic loading pattern on LLP is essential to test the durability of the LLP effectively and reliably. In this study, two simple and effective dual-peak cyclic loading test apparatuses are designed and built to evaluate the durability and functionality of LLPs. Two loading plates are used to contact the heel and toe of the prosthetic foot to simulate the dual-peak heel-strike and toe-off loadings between the foot and the ground. This dual-peak loading on the LLP is controlled and replicated by adjusting the positions of two loading plates and the actuator stroke to change the contact forces. Experimental results show that the proposed testing apparatuses and procedures can emulate the dual-peak axial loading of the LLP during normal walking gait, providing a more accurate testing method of the dynamic loading condition on LLPs than the current ISO standard.</div></div>","PeriodicalId":55381,"journal":{"name":"Biocybernetics and Biomedical Engineering","volume":"45 4","pages":"Pages 642-654"},"PeriodicalIF":6.6,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145416927","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}

引用次数: 0

Design and development of a systematic validation protocol for synthetic melanoma images for responsible use in medical artificial intelligence 设计和开发用于医疗人工智能的合成黑色素瘤图像的系统验证协议

IF 6.6 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biocybernetics and Biomedical Engineering

Pub Date : 2025-09-21 DOI: 10.1016/j.bbe.2025.09.001

Alessio Luschi , Linda Tognetti , Alessandra Cartocci , Elisa Cinotti , Giovanni Rubegni , Laura Calabrese , Martina D’onghia , Martina Dragotto , Elvira Moscarella , Gabriella Brancaccio , Giulia Briatico , Camila Scharf , Dario Buononato , Vittorio Tancredi , Carmen Cantisani , Camilla Chello , Luca Ambrosio , Pietro Scribani Rossi , Marco Virone , Giovanni Pellacani , Ernesto Iadanza

Malignant melanoma is the deadliest form of skin cancer, and artificial intelligence could help address its diagnostic challenges. Generative Adversarial Networks (GANs) can generate synthetic dermoscopic images to augment limited real datasets, but the lack of standardised validation protocols holds back models’ reliability and clinicians’ trust. This study aims to design and develop a systematic validation protocol combining quantitative metrics and qualitative expert assessments to evaluate the realism, fidelity, diversity, and usefulness of synthetic dermoscopic melanoma images. A StyleGAN2 model, designed and trained in a previous study, was selected for its superior quantitative performance and exploited to generate 25 synthetic melanoma images, matched with 25 real images. A panel of 17 dermoscopists assessed the images using a 7-point Likert scale, across multiple qualitative attributes (real vs. synthetic, skin texture, visual realism, and confidence) and pattern analysis. Accuracy, sensitivity, specificity, Fleiss’ Kappa, and Krippendorff’s Alpha were calculated to analyse inter-rater agreement and evaluation outcomes. Accuracy in real vs synthetic images classification was moderate (64 %), with sensitivity at 73 % and specificity at 56 %, with poor inter-rater concordance over qualitative attributes. Synthetic images obtained superior scores in medium visual and overall realism, and confidence level, while the frequency of recognition of pigment network-patterns was comparable with real images. The proposed holistic validation protocol can effectively estimate the quality level of synthetic dermoscopic images, regardless of the architecture of the model used for generation, offering an objective and reliable evaluation tool, as qualitative evaluations remain crucial to ensure their safe deployment in clinical settings.

恶性黑色素瘤是最致命的皮肤癌，人工智能可以帮助解决其诊断挑战。生成对抗网络（GANs）可以生成合成的皮肤镜图像来增强有限的真实数据集，但缺乏标准化的验证协议阻碍了模型的可靠性和临床医生的信任。本研究旨在设计和开发一种结合定量指标和定性专家评估的系统验证方案，以评估合成皮肤镜下黑色素瘤图像的真实感、保真度、多样性和有用性。在之前的研究中设计和训练的StyleGAN2模型因其优越的定量性能而被选中，并利用它生成25张合成黑色素瘤图像，与25张真实图像相匹配。一个由17名皮肤科医生组成的小组使用7分李克特量表评估图像，包括多个定性属性（真实与合成、皮肤纹理、视觉真实感和信心）和模式分析。计算准确性、敏感性、特异性、Fleiss Kappa和Krippendorff Alpha来分析评分者之间的一致性和评估结果。真实图像与合成图像分类的准确率为中等（64%），灵敏度为73%，特异性为56%，定性属性间一致性较差。合成图像在中等视觉和整体真实感以及置信度方面得分较高，而对色素网络模式的识别频率与真实图像相当。所提出的整体验证方案可以有效地估计合成皮肤镜图像的质量水平，而不考虑用于生成的模型的架构，提供客观可靠的评估工具，因为定性评估对于确保其在临床环境中的安全部署至关重要。

{"title":"Design and development of a systematic validation protocol for synthetic melanoma images for responsible use in medical artificial intelligence","authors":"Alessio Luschi , Linda Tognetti , Alessandra Cartocci , Elisa Cinotti , Giovanni Rubegni , Laura Calabrese , Martina D’onghia , Martina Dragotto , Elvira Moscarella , Gabriella Brancaccio , Giulia Briatico , Camila Scharf , Dario Buononato , Vittorio Tancredi , Carmen Cantisani , Camilla Chello , Luca Ambrosio , Pietro Scribani Rossi , Marco Virone , Giovanni Pellacani , Ernesto Iadanza","doi":"10.1016/j.bbe.2025.09.001","DOIUrl":"10.1016/j.bbe.2025.09.001","url":null,"abstract":"<div><div>Malignant melanoma is the deadliest form of skin cancer, and artificial intelligence could help address its diagnostic challenges. Generative Adversarial Networks (GANs) can generate synthetic dermoscopic images to augment limited real datasets, but the lack of standardised validation protocols holds back models’ reliability and clinicians’ trust. This study aims to design and develop a systematic validation protocol combining quantitative metrics and qualitative expert assessments to evaluate the realism, fidelity, diversity, and usefulness of synthetic dermoscopic melanoma images. A StyleGAN2 model, designed and trained in a previous study, was selected for its superior quantitative performance and exploited to generate 25 synthetic melanoma images, matched with 25 real images. A panel of 17 dermoscopists assessed the images using a 7-point Likert scale, across multiple qualitative attributes (real vs. synthetic, skin texture, visual realism, and confidence) and pattern analysis. Accuracy, sensitivity, specificity, Fleiss’ Kappa, and Krippendorff’s Alpha were calculated to analyse inter-rater agreement and evaluation outcomes. Accuracy in real vs synthetic images classification was moderate (64 %), with sensitivity at 73 % and specificity at 56 %, with poor inter-rater concordance over qualitative attributes. Synthetic images obtained superior scores in medium visual and overall realism, and confidence level, while the frequency of recognition of pigment network-patterns was comparable with real images. The proposed holistic validation protocol can effectively estimate the quality level of synthetic dermoscopic images, regardless of the architecture of the model used for generation, offering an objective and reliable evaluation tool, as qualitative evaluations remain crucial to ensure their safe deployment in clinical settings.</div></div>","PeriodicalId":55381,"journal":{"name":"Biocybernetics and Biomedical Engineering","volume":"45 4","pages":"Pages 608-616"},"PeriodicalIF":6.6,"publicationDate":"2025-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145099872","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}

引用次数: 0