Biomedical Signal Processing and Control最新文献_第7页

Development of a wearable ultrasound–FES integrated rehabilitation and motor-functional reconstruction system for post-stroke patients 为脑卒中后患者开发可穿戴超声-FES 集成康复和运动功能重建系统

IF 4.9 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biomedical Signal Processing and Control

Pub Date : 2024-11-02 DOI: 10.1016/j.bspc.2024.106846

Yudong Cao , Yun Lu , Wenpan Wang , Peng Xu , Xiaoli Yang , Shiwu Zhang , Ming Wu , Xinglong Gong , Shuaishuai Sun

Post-stroke patients experience a significant decrease of self-care capabilities in their daily lives because of motor dysfunction. The combination of intention recognition and functional electrical stimulation (FES) is used frequently to assist in improving the self-care capabilities for post-stroke patients. However, the electrical noise from the environment and the weak bio-signal from post-stroke patients lead to low-accurate intention recognition for post-stroke patients. To overcome the issue, this paper introduces a wearable rehabilitation and motor-functional reconstruction system for post-stroke rehabilitation with a new intention recognition system. This system consists of an FES unit and a wearable musculoskeletal ultrasound system. The integration of the wearable ultrasound system allows for high-accuracy continuous intention recognition whilst the FES unit is in operation. This key feature significantly enhances the system’s robustness in FES control, augments the signal-to-noise ratio and offers precise assistance in the reconstruction of motor function, thereby improving the effectiveness of post-stroke rehabilitation. In this study, the feasibility and efficiency of the proposed system were investigated. In the clinical trial, eight post-stroke subjects were recruited. In the experiment of motor-functional reconstruction, the proposed system demonstrated enhancements of approximately 23 % and 76 % in wrist raising angle and velocity, respectively. These results demonstrated that the proposed wearable system is effective for active rehabilitation and potential candidate to reconstruct the motor function of post-stroke patients.

由于运动功能障碍，脑卒中后患者的日常生活自理能力明显下降。意向识别和功能性电刺激（FES）的结合被广泛用于帮助改善脑卒中后患者的自理能力。然而，来自环境的电噪声和脑卒中后患者微弱的生物信号导致脑卒中后患者的意向识别准确率较低。为了克服这一问题，本文介绍了一种用于中风后康复的可穿戴康复和运动功能重建系统，其中包含一个新的意向识别系统。该系统由 FES 单元和可穿戴式肌肉骨骼超声系统组成。可穿戴式超声波系统的集成，使其能够在 FES 装置运行时进行高精度的连续意向识别。这一关键功能大大增强了系统在 FES 控制中的稳健性，提高了信噪比，为运动功能的重建提供了精确的帮助，从而提高了中风后康复的效果。本研究调查了拟议系统的可行性和效率。在临床试验中，共招募了 8 名中风后受试者。在运动功能重建实验中，拟议的系统在手腕抬起角度和速度方面分别提高了约 23% 和 76%。这些结果表明，建议的可穿戴系统对主动康复有效，是重建中风后患者运动功能的潜在候选系统。

{"title":"Development of a wearable ultrasound–FES integrated rehabilitation and motor-functional reconstruction system for post-stroke patients","authors":"Yudong Cao , Yun Lu , Wenpan Wang , Peng Xu , Xiaoli Yang , Shiwu Zhang , Ming Wu , Xinglong Gong , Shuaishuai Sun","doi":"10.1016/j.bspc.2024.106846","DOIUrl":"10.1016/j.bspc.2024.106846","url":null,"abstract":"<div><div>Post-stroke patients experience a significant decrease of self-care capabilities in their daily lives because of motor dysfunction. The combination of intention recognition and functional electrical stimulation (FES) is used frequently to assist in improving the self-care capabilities for post-stroke patients. However, the electrical noise from the environment and the weak bio-signal from post-stroke patients lead to low-accurate intention recognition for post-stroke patients. To overcome the issue, this paper introduces a wearable rehabilitation and motor-functional reconstruction system for post-stroke rehabilitation with a new intention recognition system. This system consists of an FES unit and a wearable musculoskeletal ultrasound system. The integration of the wearable ultrasound system allows for high-accuracy continuous intention recognition whilst the FES unit is in operation. This key feature significantly enhances the system’s robustness in FES control, augments the signal-to-noise ratio and offers precise assistance in the reconstruction of motor function, thereby improving the effectiveness of post-stroke rehabilitation. In this study, the feasibility and efficiency of the proposed system were investigated. In the clinical trial, eight post-stroke subjects were recruited. In the experiment of motor-functional reconstruction, the proposed system demonstrated enhancements of approximately 23 % and 76 % in wrist raising angle and velocity, respectively. These results demonstrated that the proposed wearable system is effective for active rehabilitation and potential candidate to reconstruct the motor function of post-stroke patients.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"100 ","pages":"Article 106846"},"PeriodicalIF":4.9,"publicationDate":"2024-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142572445","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

ROSE-Net: Leveraging remote photoplethysmography to estimate oxygen saturation using deep learning ROSE-Net：利用深度学习估算血氧饱和度

IF 4.9 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biomedical Signal Processing and Control

Pub Date : 2024-11-01 DOI: 10.1016/j.bspc.2024.107105

Moajjem Hossain Chowdhury , Mamun Bin Ibne Reaz , Sawal Hamid Md Ali , Muhammad Salman Khan , Muhammad E.H. Chowdhury

A method for accurately estimating physiological signals from video streams at a minimal cost holds immense value, particularly in pre-clinical health monitoring applications. This technique is particularly indispensable in scenarios where traditional sensors, such as finger photoplethysmography (PPG), are not viable, such as in cases involving burn victims, premature infants, or individuals with sensitive skin. Remote photoplethysmography (rPPG) is a process of estimating PPG signals using video streams instead of traditional sensors. rPPG has thus been seen as a promising alternative to traditional PPG. As an alternative to using PPG for estimating oxygen saturation (SpO2), we propose ROSE-Net. ROSE-Net, trained on clinical PPG, was tested on an external rPPG dataset, PURE. The model achieved a mean absolute error (MAE) of 1.20 and a root mean square error (RMSE) of 1.86 on clinical PPG. When tested on rPPG, it exhibited an MAE of 1.95 and an RMSE of 2.46 in PURE, an MAE of 0.77, and an RMSE of 0.96 in ARPOS. These results demonstrate the model’s ability to estimate SpO2 levels within acceptable margins when applied to rPPG data. Consequently, rPPG presents a viable approach for estimating SpO2 levels, paving the way for non-contact health tracking applications.

一种能以最低成本从视频流中准确估计生理信号的方法具有巨大价值，尤其是在临床前健康监测应用中。这种技术在传统传感器（如手指血压计）无法使用的情况下尤其不可或缺，例如在涉及烧伤患者、早产儿或皮肤敏感的人的情况下。因此，rPPG 被视为传统 PPG 的一种有前途的替代方法。作为使用 PPG 估算血氧饱和度（SpO2）的替代方法，我们提出了 ROSE-Net 方案。ROSE-Net 以临床 PPG 为基础进行训练，并在外部 rPPG 数据集 PURE 上进行了测试。该模型在临床 PPG 上的平均绝对误差 (MAE) 为 1.20，均方根误差 (RMSE) 为 1.86。在 rPPG 上进行测试时，PURE 的 MAE 为 1.95，RMSE 为 2.46；ARPOS 的 MAE 为 0.77，RMSE 为 0.96。这些结果表明，该模型在应用于 rPPG 数据时，能够在可接受的范围内估计 SpO2 水平。因此，rPPG 是估计 SpO2 水平的一种可行方法，为非接触式健康跟踪应用铺平了道路。

{"title":"ROSE-Net: Leveraging remote photoplethysmography to estimate oxygen saturation using deep learning","authors":"Moajjem Hossain Chowdhury , Mamun Bin Ibne Reaz , Sawal Hamid Md Ali , Muhammad Salman Khan , Muhammad E.H. Chowdhury","doi":"10.1016/j.bspc.2024.107105","DOIUrl":"10.1016/j.bspc.2024.107105","url":null,"abstract":"<div><div>A method for accurately estimating physiological signals from video streams at a minimal cost holds immense value, particularly in pre-clinical health monitoring applications. This technique is particularly indispensable in scenarios where traditional sensors, such as finger photoplethysmography (PPG), are not viable, such as in cases involving burn victims, premature infants, or individuals with sensitive skin. Remote photoplethysmography (rPPG) is a process of estimating PPG signals using video streams instead of traditional sensors. rPPG has thus been seen as a promising alternative to traditional PPG. As an alternative to using PPG for estimating oxygen saturation (SpO2), we propose ROSE-Net. ROSE-Net, trained on clinical PPG, was tested on an external rPPG dataset, PURE. The model achieved a mean absolute error (MAE) of 1.20 and a root mean square error (RMSE) of 1.86 on clinical PPG. When tested on rPPG, it exhibited an MAE of 1.95 and an RMSE of 2.46 in PURE, an MAE of 0.77, and an RMSE of 0.96 in ARPOS. These results demonstrate the model’s ability to estimate SpO2 levels within acceptable margins when applied to rPPG data. Consequently, rPPG presents a viable approach for estimating SpO2 levels, paving the way for non-contact health tracking applications.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"100 ","pages":"Article 107105"},"PeriodicalIF":4.9,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142572448","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

MindCeive: Perceiving human imagination using CNN-GRU and GANs MindCeive：利用 CNN-GRU 和 GAN 感知人类的想象力

IF 4.9 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biomedical Signal Processing and Control

Pub Date : 2024-11-01 DOI: 10.1016/j.bspc.2024.107110

Ritik Naik, Kunal Chaudhari, Ketaki Jadhav, Amit Joshi

Neuroscience has made astonishing advancements in understanding the human brain with the help of Brain-Computer Interface. Recent contributions in the field of Artificial Intelligence by different researchers made it possible to perceive human imagination by decoding brain signals. Generating visual stimuli perceived by humans will help in analyzing how the human brain works and behaves to different perceptual experiences. Different techniques like Electroencephalography, Magnetoencephalography, functional Magnetic Resonance Imaging, etc. are used to capture brain signals. Electroencephalography signals are non-invasive, low cost, and also have high temporal resolution, therefore they are preferred. Machine learning models are used to extract important features from these signals. These extracted features are then used by Generative Adversarial Network to generate images representing human imagination. This work uses Electroencephalography signals to generate realistic images. The task of extracting important features from Electroencephalography signals is achieved using Convolutional Neural Network and Gated Recurrent Unit based feature extractor. The proposed feature extractor accomplishes better classification accuracy than existing models. By using these extracted features in combination with proposed novel architecture of Generative Adversarial Network, realistic images of objects imagined by humans are generated. The proposed MindCeive approach outperforms previous works by showing improvement in various performance metrics such as Classification Accuracy, Inception Score, and Class Diversity Score.

在脑机接口的帮助下，神经科学在了解人类大脑方面取得了惊人的进步。最近，不同研究人员在人工智能领域做出的贡献使得通过解码大脑信号来感知人类的想象力成为可能。生成人类感知到的视觉刺激将有助于分析人脑是如何工作的以及在不同的感知体验中是如何表现的。脑电图、脑磁图、功能磁共振成像等不同技术被用来捕捉大脑信号。脑电信号无创、成本低、时间分辨率高，因此是首选。机器学习模型用于从这些信号中提取重要特征。然后，生成对抗网络利用这些提取的特征生成代表人类想象力的图像。这项工作使用脑电信号生成逼真的图像。从脑电信号中提取重要特征的任务是通过基于卷积神经网络和门控递归单元的特征提取器来完成的。与现有模型相比，所提出的特征提取器的分类准确性更高。通过将这些提取的特征与所提出的生成对抗网络新架构相结合，可以生成人类所想象物体的逼真图像。所提出的 MindCeive 方法在分类准确率、起始得分和类别多样性得分等各种性能指标上都有所改进，因而优于之前的研究成果。

{"title":"MindCeive: Perceiving human imagination using CNN-GRU and GANs","authors":"Ritik Naik, Kunal Chaudhari, Ketaki Jadhav, Amit Joshi","doi":"10.1016/j.bspc.2024.107110","DOIUrl":"10.1016/j.bspc.2024.107110","url":null,"abstract":"<div><div>Neuroscience has made astonishing advancements in understanding the human brain with the help of Brain-Computer Interface. Recent contributions in the field of Artificial Intelligence by different researchers made it possible to perceive human imagination by decoding brain signals. Generating visual stimuli perceived by humans will help in analyzing how the human brain works and behaves to different perceptual experiences. Different techniques like Electroencephalography, Magnetoencephalography, functional Magnetic Resonance Imaging, etc. are used to capture brain signals. Electroencephalography signals are non-invasive, low cost, and also have high temporal resolution, therefore they are preferred. Machine learning models are used to extract important features from these signals. These extracted features are then used by Generative Adversarial Network to generate images representing human imagination. This work uses Electroencephalography signals to generate realistic images. The task of extracting important features from Electroencephalography signals is achieved using Convolutional Neural Network and Gated Recurrent Unit based feature extractor. The proposed feature extractor accomplishes better classification accuracy than existing models. By using these extracted features in combination with proposed novel architecture of Generative Adversarial Network, realistic images of objects imagined by humans are generated. The proposed MindCeive approach outperforms previous works by showing improvement in various performance metrics such as Classification Accuracy, Inception Score, and Class Diversity Score.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"100 ","pages":"Article 107110"},"PeriodicalIF":4.9,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142573359","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

Coarse-to-Fine bone age regression by using multi-scale self-attention mechanism 利用多尺度自我关注机制进行粗-细骨龄回归

IF 4.9 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biomedical Signal Processing and Control

Pub Date : 2024-11-01 DOI: 10.1016/j.bspc.2024.107029

Guanyu Wu , Ziming Wang , Jian Peng , Shaobing Gao

Pediatric bone age assessment (BAA) is a widely-used clinical technique employed to investigate various growth, genetic, and endocrine disorders in children. In this article, we propose a novel network architecture called BoGFF-Net that integrates multi-scale hand bone feature maps across different levels, and introduce an adaptive triplet loss (ATL) function that can distinguish sample pairs in regression tasks. Our network incorporates self-attention mechanisms to adaptively learn the most important regions of hand bone images, which deviates from the conventional approach of extracting specific regions in the field of bone age assessment. Additionally, we observe heterogeneous characteristics of hand bone development among different age ranges in adolescents. Therefore, we introduce a two-stage coarse-to-fine framework that can accommodate greater differences in bone modalities across diverse age groups. Quantitative and qualitative results from extensive experiments on two public bone age datasets highlight the performance and effectiveness of our model. Specifically, our model achieves competitive performance with a 3.91 mean absolute error (MAE) on the RSNA test dataset, compared to the latest model proposed by Yang et al. 2023, and a 7.07 MAE on the DHA dataset, setting a new state-of-the-art benchmark. The data and code are available at: BoGFF-Net

小儿骨龄评估（BAA）是一项广泛应用的临床技术，用于研究儿童的各种生长、遗传和内分泌疾病。在本文中，我们提出了一种名为 BoGFF-Net 的新型网络架构，它整合了不同层次的多尺度手骨特征图，并引入了一种自适应三重损失（ATL）函数，可在回归任务中区分样本对。我们的网络结合了自我注意机制，能够自适应地学习手骨图像中最重要的区域，这与骨龄评估领域提取特定区域的传统方法不同。此外，我们还观察到不同年龄段青少年手骨发育的异质性特征。因此，我们引入了一个从粗到细的两阶段框架，以适应不同年龄段骨骼模式的更大差异。在两个公共骨龄数据集上进行的大量实验得出的定量和定性结果，凸显了我们模型的性能和有效性。具体来说，与 Yang 等人提出的最新模型相比，我们的模型在 RSNA 测试数据集上取得了 3.91 的平均绝对误差（MAE），在 DHA 数据集上取得了 7.07 的平均绝对误差（MAE），树立了新的先进基准。数据和代码见BoGFF-Net

{"title":"Coarse-to-Fine bone age regression by using multi-scale self-attention mechanism","authors":"Guanyu Wu , Ziming Wang , Jian Peng , Shaobing Gao","doi":"10.1016/j.bspc.2024.107029","DOIUrl":"10.1016/j.bspc.2024.107029","url":null,"abstract":"<div><div>Pediatric bone age assessment (BAA) is a widely-used clinical technique employed to investigate various growth, genetic, and endocrine disorders in children. In this article, we propose a novel network architecture called BoGFF-Net that integrates multi-scale hand bone feature maps across different levels, and introduce an adaptive triplet loss (ATL) function that can distinguish sample pairs in regression tasks. Our network incorporates self-attention mechanisms to adaptively learn the most important regions of hand bone images, which deviates from the conventional approach of extracting specific regions in the field of bone age assessment. Additionally, we observe heterogeneous characteristics of hand bone development among different age ranges in adolescents. Therefore, we introduce a two-stage coarse-to-fine framework that can accommodate greater differences in bone modalities across diverse age groups. Quantitative and qualitative results from extensive experiments on two public bone age datasets highlight the performance and effectiveness of our model. Specifically, our model achieves competitive performance with a 3.91 mean absolute error (MAE) on the RSNA test dataset, compared to the latest model proposed by Yang et al. 2023, and a 7.07 MAE on the DHA dataset, setting a new state-of-the-art benchmark. The data and code are available at: <span><span>BoGFF-Net</span><svg><path></path></svg></span></div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"100 ","pages":"Article 107029"},"PeriodicalIF":4.9,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142573360","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

Medical image fusion via decoupled representation and component-wise regularization learning 通过解耦表示和分量正则化学习实现医学图像融合

IF 4.9 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biomedical Signal Processing and Control

Pub Date : 2024-11-01 DOI: 10.1016/j.bspc.2024.106859

Rui Zhang , Haoze Sun , Lizhen Deng , Hu Zhu , Wei Qian

Medical image fusion plays an important role in the precise diagnosis, treatment planning, and follow-up studies of various diseases. While tremendous improvements in medical image fusion based on convolution sparse coding have been achieved, existing methods are still limited by the intractable redundancy information interaction between source medical images. In this paper, we propose an easy yet effective representation and regularization learning method based on decomposed components scheme with high competitive performance. We construct more compact information interactions by decoupled representation learning, which simultaneously mitigates the problem of redundancy in fusion component entanglement. And then two different regularization operators are adaptively exploited to depict two different components separately, which describe the structural-inspired difference based on the decoupled principle. Furthermore, we combine the alternating direction method of multipliers (ADMM) algorithm and the conjugate gradient (CG) method to optimize our proposed model. Our experiments demonstrate that our proposed method has significant improvements in efficiency and fusion performance against the state-of-the-art methods.

医学图像融合在各种疾病的精确诊断、治疗计划和随访研究中发挥着重要作用。虽然基于卷积稀疏编码的医学图像融合取得了巨大进步，但现有方法仍受限于源医学图像之间难以解决的冗余信息交互问题。在本文中，我们提出了一种基于分解组件方案的简单而有效的表示和正则化学习方法，具有很高的性能竞争力。我们通过解耦表示学习构建了更紧凑的信息交互，同时缓解了融合组件纠缠中的冗余问题。然后，我们自适应地利用两种不同的正则化算子来分别描述两个不同的分量，从而描述了基于解耦原理的结构启发式差异。此外，我们还结合了交替乘法（ADMM）算法和共轭梯度（CG）方法来优化我们提出的模型。实验证明，与最先进的方法相比，我们提出的方法在效率和融合性能方面都有显著提高。

{"title":"Medical image fusion via decoupled representation and component-wise regularization learning","authors":"Rui Zhang , Haoze Sun , Lizhen Deng , Hu Zhu , Wei Qian","doi":"10.1016/j.bspc.2024.106859","DOIUrl":"10.1016/j.bspc.2024.106859","url":null,"abstract":"<div><div>Medical image fusion plays an important role in the precise diagnosis, treatment planning, and follow-up studies of various diseases. While tremendous improvements in medical image fusion based on convolution sparse coding have been achieved, existing methods are still limited by the intractable redundancy information interaction between source medical images. In this paper, we propose an easy yet effective representation and regularization learning method based on decomposed components scheme with high competitive performance. We construct more compact information interactions by decoupled representation learning, which simultaneously mitigates the problem of redundancy in fusion component entanglement. And then two different regularization operators are adaptively exploited to depict two different components separately, which describe the structural-inspired difference based on the decoupled principle. Furthermore, we combine the alternating direction method of multipliers (ADMM) algorithm and the conjugate gradient (CG) method to optimize our proposed model. Our experiments demonstrate that our proposed method has significant improvements in efficiency and fusion performance against the state-of-the-art methods.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"100 ","pages":"Article 106859"},"PeriodicalIF":4.9,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142572444","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

PhysKANNet: A KAN-based model for multiscale feature extraction and contextual fusion in remote physiological measurement PhysKANNet：基于 KAN 的远程生理测量多尺度特征提取和上下文融合模型

IF 4.9 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biomedical Signal Processing and Control

Pub Date : 2024-11-01 DOI: 10.1016/j.bspc.2024.107111

Tianqi Liu , Hanguang Xiao , Yisha Sun , Kun Zuo , Zhipeng Li , Zhiying Yang , Shihong Liu

Physiological indicator reflects the health status of the human body, and remote photoplethysmography (rPPG) is a highly promising technology for contactless measurement of these indicators through facial video. However, current deep learning methods mainly rely on traditional neural networks with limited spatiotemporal receptive fields, overlooking the importance of multi-scale features and noise resistance in rPPG signal modeling. This results in challenges when addressing subtle color changes and noise interference. To overcome these limitations, we leverage the advantages of the Kolmogorov-Arnold Network (KAN) in handling sparse data and propose PhysKANNet, a novel KAN-based encoder–decoder architecture that integrates multi-scale feature extraction and contextual information fusion to enhance rPPG signal extraction. We introduce three new plug-and-play modules for PhysKANNet: the rPPG-Aware Convolutional Attention Block, which extracts features at different scales through a multi-branch structure and enhances multi-scale representation using KAN’s nonlinear modeling capabilities; the Multi-Dimensional Feature Fusion Block, which combines high-dimensional features from the encoder with low-dimensional features from the decoder; and the rPPG Edge Sampling Block, which fuses edge and semantic information to further optimize signal extraction accuracy. We employ unsupervised learning for training PhysKANNet and conducted comprehensive experiments on multiple benchmark datasets. The results show that PhysKANNet significantly improves feature learning from unlabeled data, achieving excellent performance across various testing scenarios.

生理指标反映了人体的健康状况，而远程血压计（rPPG）是一种通过面部视频进行非接触式测量的极具前景的技术。然而，目前的深度学习方法主要依赖于时空感受野有限的传统神经网络，忽视了多尺度特征和抗噪性在 rPPG 信号建模中的重要性。这导致在处理微妙的颜色变化和噪声干扰时面临挑战。为了克服这些局限性，我们利用柯尔莫哥洛夫-阿诺德网络（KAN）在处理稀疏数据方面的优势，提出了基于 KAN 的新型编码器-解码器架构 PhysKANNet，该架构集成了多尺度特征提取和上下文信息融合，以增强 rPPG 信号提取。我们为 PhysKANNet 引入了三个新的即插即用模块：rPPG 感知卷积注意力模块（rPPG-Aware Convolutional Attention Block），它通过多分支结构提取不同尺度的特征，并利用 KAN 的非线性建模能力增强多尺度表示；多维特征融合模块（Multi-Dimensional Feature Fusion Block），它将来自编码器的高维特征与来自解码器的低维特征相结合；以及 rPPG 边缘采样模块（rPPG Edge Sampling Block），它融合边缘和语义信息，进一步优化信号提取的准确性。我们采用无监督学习来训练 PhysKANNet，并在多个基准数据集上进行了综合实验。结果表明，PhysKANNet 显著提高了无标记数据的特征学习能力，在各种测试场景中都取得了优异的性能。

{"title":"PhysKANNet: A KAN-based model for multiscale feature extraction and contextual fusion in remote physiological measurement","authors":"Tianqi Liu , Hanguang Xiao , Yisha Sun , Kun Zuo , Zhipeng Li , Zhiying Yang , Shihong Liu","doi":"10.1016/j.bspc.2024.107111","DOIUrl":"10.1016/j.bspc.2024.107111","url":null,"abstract":"<div><div>Physiological indicator reflects the health status of the human body, and remote photoplethysmography (rPPG) is a highly promising technology for contactless measurement of these indicators through facial video. However, current deep learning methods mainly rely on traditional neural networks with limited spatiotemporal receptive fields, overlooking the importance of multi-scale features and noise resistance in rPPG signal modeling. This results in challenges when addressing subtle color changes and noise interference. To overcome these limitations, we leverage the advantages of the Kolmogorov-Arnold Network (KAN) in handling sparse data and propose PhysKANNet, a novel KAN-based encoder–decoder architecture that integrates multi-scale feature extraction and contextual information fusion to enhance rPPG signal extraction. We introduce three new plug-and-play modules for PhysKANNet: the rPPG-Aware Convolutional Attention Block, which extracts features at different scales through a multi-branch structure and enhances multi-scale representation using KAN’s nonlinear modeling capabilities; the Multi-Dimensional Feature Fusion Block, which combines high-dimensional features from the encoder with low-dimensional features from the decoder; and the rPPG Edge Sampling Block, which fuses edge and semantic information to further optimize signal extraction accuracy. We employ unsupervised learning for training PhysKANNet and conducted comprehensive experiments on multiple benchmark datasets. The results show that PhysKANNet significantly improves feature learning from unlabeled data, achieving excellent performance across various testing scenarios.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"100 ","pages":"Article 107111"},"PeriodicalIF":4.9,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142572447","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

Liver disease classification using histogram-based gradient boosting classification tree with feature selection algorithm 利用基于直方图的梯度提升分类树和特征选择算法进行肝病分类

IF 4.9 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biomedical Signal Processing and Control

Pub Date : 2024-10-31 DOI: 10.1016/j.bspc.2024.107102

Prasannavenkatesan Theerthagiri

Healthcare is the key for everyone to run daily life, and health diagnosing techniques should be accessible easily. Indeed, the early identification of liver disease will be supportive for physicians to make decisions. Utilizing feature selection and classification approaches, this work aims to predict liver disorders through machine learning. The Histogram-based Gradient Boosting Classification Tree with a recursive feature selection algorithm (HGBoost) is proposed in this paper. The recursive feature selection approach and the Gradient Boosting are used to forecast liver disease. Using data from Indian liver patient records, the proposed HGBoost method has been assessed. Assessing the accuracy, confusion matrix, and area under curve involves implementing and comparing a variety of classification techniques, including MLP, Gboost, Adaboost, and proposed HGBoost algorithms. With the help of the recursive feature selection technique, the proposed HGBoost has surpassed other current algorithms. In comparison to the MLP, RF, Gboost, Adaboost, and proposed HGBoost algorithms, the enhanced accuracy is between 4 and 9% and between 1 and 7 % of the MSE error.

医疗保健是每个人日常生活的关键，而健康诊断技术应易于获取。事实上，肝脏疾病的早期识别将有助于医生做出决策。本研究利用特征选择和分类方法，旨在通过机器学习预测肝脏疾病。本文提出了基于直方图梯度提升分类树的递归特征选择算法（HGBoost）。递归特征选择方法和梯度提升用于预测肝病。利用印度肝病患者的记录数据，对所提出的 HGBoost 方法进行了评估。评估准确率、混淆矩阵和曲线下面积涉及实施和比较各种分类技术，包括 MLP、Gboost、Adaboost 和拟议的 HGBoost 算法。在递归特征选择技术的帮助下，拟议的 HGBoost 算法超越了其他现有算法。与 MLP、RF、Gboost、Adaboost 和提议的 HGBoost 算法相比，准确率提高了 4% 到 9%，MSE 误差提高了 1% 到 7%。

{"title":"Liver disease classification using histogram-based gradient boosting classification tree with feature selection algorithm","authors":"Prasannavenkatesan Theerthagiri","doi":"10.1016/j.bspc.2024.107102","DOIUrl":"10.1016/j.bspc.2024.107102","url":null,"abstract":"<div><div>Healthcare is the key for everyone to run daily life, and health diagnosing techniques should be accessible easily. Indeed, the early identification of liver disease will be supportive for physicians to make decisions. Utilizing feature selection and classification approaches, this work aims to predict liver disorders through machine learning. The Histogram-based Gradient Boosting Classification Tree with a recursive feature selection algorithm (HGBoost) is proposed in this paper. The recursive feature selection approach and the Gradient Boosting are used to forecast liver disease. Using data from Indian liver patient records, the proposed HGBoost method has been assessed. Assessing the accuracy, confusion matrix, and area under curve involves implementing and comparing a variety of classification techniques, including MLP, Gboost, Adaboost, and proposed HGBoost algorithms. With the help of the recursive feature selection technique, the proposed HGBoost has surpassed other current algorithms. In comparison to the MLP, RF, Gboost, Adaboost, and proposed HGBoost algorithms, the enhanced accuracy is between 4 and 9% and between 1 and 7 % of the MSE error.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"100 ","pages":"Article 107102"},"PeriodicalIF":4.9,"publicationDate":"2024-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142561152","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

Video-based heart rate estimation with spectrogram signal quality ranking and fusion 基于视频的心率估计与频谱图信号质量排序和融合

IF 4.9 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biomedical Signal Processing and Control

Pub Date : 2024-10-30 DOI: 10.1016/j.bspc.2024.107094

Rencheng Song , Zhenzhou Du , Juan Cheng , Chang Li , Xuezhi Yang

Remote photoplethysmography (rPPG) enables non-contact measurement of heart rate (HR). However, the stability of rPPG extraction is a bottleneck limiting its application. To address this issue, a signal quality ranking and fusion (SQRF) approach based on HR continuity in the time–frequency domain is introduced. Firstly, the facial region is divided into multiple regions of interest (ROIs), and the raw blood volume pulse (BVP) signal is extracted from each ROI separately using a conventional rPPG method such as the plane orthogonal to skin (POS) method. Then, wavelet synchrosqueezed transform (WSST) is employed to convert the raw pulse signals into spectrograms, which are further ranked according to the HR instantaneous continuity. The selected spectrograms with high-quality HR continuity are then fused using a weighted average to predict the final HR. The proposed SQRF algorithm is verified on three public datasets DDPM, UBFC-Phys and PURE with real scenarios. The obtained mean absolute error (MAE) was reduced by 58.7%, 47.5%, and 16.0% respectively, compared to the original single-ROI method. The results prove that SORF with spectrogram-based HR continuity can consistently boost the stability of POS.

远程心电图（rPPG）可以非接触式测量心率（HR）。然而，rPPG 提取的稳定性是限制其应用的瓶颈。为了解决这个问题，我们引入了一种基于时频域心率连续性的信号质量排序和融合（SQRF）方法。首先，将面部区域划分为多个感兴趣区域（ROI），然后使用传统的 rPPG 方法（如与皮肤正交的平面（POS）方法）分别从每个感兴趣区域提取原始血容量脉搏（BVP）信号。然后，采用小波同步萃取变换（WSST）将原始脉搏信号转换为频谱图，并根据心率瞬时连续性对频谱图进一步排序。然后使用加权平均法融合所选的具有高质量心率连续性的频谱图，以预测最终心率。提议的 SQRF 算法在三个公共数据集 DDPM、UBFC-Phys 和 PURE 上进行了真实场景验证。与原始的单 ROI 方法相比，获得的平均绝对误差（MAE）分别减少了 58.7%、47.5% 和 16.0%。结果证明，基于频谱图 HR 连续性的 SORF 可以持续提高 POS 的稳定性。

{"title":"Video-based heart rate estimation with spectrogram signal quality ranking and fusion","authors":"Rencheng Song , Zhenzhou Du , Juan Cheng , Chang Li , Xuezhi Yang","doi":"10.1016/j.bspc.2024.107094","DOIUrl":"10.1016/j.bspc.2024.107094","url":null,"abstract":"<div><div>Remote photoplethysmography (rPPG) enables non-contact measurement of heart rate (HR). However, the stability of rPPG extraction is a bottleneck limiting its application. To address this issue, a signal quality ranking and fusion (SQRF) approach based on HR continuity in the time–frequency domain is introduced. Firstly, the facial region is divided into multiple regions of interest (ROIs), and the raw blood volume pulse (BVP) signal is extracted from each ROI separately using a conventional rPPG method such as the plane orthogonal to skin (POS) method. Then, wavelet synchrosqueezed transform (WSST) is employed to convert the raw pulse signals into spectrograms, which are further ranked according to the HR instantaneous continuity. The selected spectrograms with high-quality HR continuity are then fused using a weighted average to predict the final HR. The proposed SQRF algorithm is verified on three public datasets DDPM, UBFC-Phys and PURE with real scenarios. The obtained mean absolute error (MAE) was reduced by 58.7%, 47.5%, and 16.0% respectively, compared to the original single-ROI method. The results prove that SORF with spectrogram-based HR continuity can consistently boost the stability of POS.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"100 ","pages":"Article 107094"},"PeriodicalIF":4.9,"publicationDate":"2024-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142551984","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

CoLM: Contrastive learning and multiple instance learning network for lung cancer classification of surgical options based on frozen pathological images CoLM：基于冰冻病理图像的肺癌手术方案分类对比学习和多实例学习网络

IF 4.9 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biomedical Signal Processing and Control

Pub Date : 2024-10-30 DOI: 10.1016/j.bspc.2024.107097

Lu Zhao , Wangyuan Zhao , Lu Qiu , Mengqi Jiang , Liqiang Qian , Hua-Nong Ting , Xiaolong Fu , Puming Zhang , Yuchen Han , Jun Zhao

Histopathological images are regarded as the gold standard in cancer diagnosis. Formalin-fixed paraffin-embedded (FFPE) tissues are routinely collected and archived for pathological examination. However, the time-consuming procedures of tissue fixation and embedding render FFPE tissues unsuitable for intraoperative diagnosis, where immediate results are crucial during surgical procedures. In contrast, obtaining a fresh frozen section (FS) takes a very short time. FS samples are widely utilized for intraoperative diagnosis, whereas the diagnostic accuracy of FS is currently limited by the presence of potential histological artifacts. In this paper, we propose a contrastive learning image translation and multiple instance learning network (CoLM) for lung cancer classification. CoLM efficiently translates FS images into FFPE-style images and facilitates whole slide image classification. The entire framework encompasses two crucial stages. In the first stage, we employ a contrastive learning translation network with a dual-attention module (CL-DAM) for image translation. In the second stage, we utilize a hybrid transformer multi-instance learning-based network (HTM) to address the challenge posed by weak labels. We conduct experiments on lung cancer datasets to validate the performance of our proposed approach. The results demonstrate that our method achieve superior classification performance over other state-of-the-art methods, effectively mitigating the impact of blurred FS images. The proposed framework not only elevates the precision of intraoperative diagnosis when employing FS but also provides valuable reference for pathologists through the application of synthetic images.

组织病理学图像被视为癌症诊断的黄金标准。福尔马林固定石蜡包埋（FFPE）组织被常规收集和存档，用于病理检查。然而，由于组织固定和包埋过程耗时，FFPE 组织不适合用于术中诊断，因为在手术过程中，即时结果至关重要。相比之下，获取新鲜冷冻切片（FS）只需很短的时间。新鲜冰冻切片样本被广泛用于术中诊断，但由于存在潜在的组织学伪影，目前新鲜冰冻切片的诊断准确性受到限制。在本文中，我们提出了一种用于肺癌分类的对比学习图像翻译和多实例学习网络（CoLM）。CoLM 能有效地将 FS 图像转换为 FFPE 类型的图像，并促进整张切片图像的分类。整个框架包括两个关键阶段。在第一阶段，我们采用带有双注意模块（CL-DAM）的对比学习翻译网络进行图像翻译。在第二阶段，我们利用基于多实例学习的混合转换器网络（HTM）来应对弱标签带来的挑战。我们在肺癌数据集上进行了实验，以验证我们提出的方法的性能。结果表明，与其他最先进的方法相比，我们的方法实现了更优越的分类性能，有效减轻了模糊 FS 图像的影响。所提出的框架不仅提高了使用 FS 时术中诊断的精确度，还通过合成图像的应用为病理学家提供了有价值的参考。

{"title":"CoLM: Contrastive learning and multiple instance learning network for lung cancer classification of surgical options based on frozen pathological images","authors":"Lu Zhao , Wangyuan Zhao , Lu Qiu , Mengqi Jiang , Liqiang Qian , Hua-Nong Ting , Xiaolong Fu , Puming Zhang , Yuchen Han , Jun Zhao","doi":"10.1016/j.bspc.2024.107097","DOIUrl":"10.1016/j.bspc.2024.107097","url":null,"abstract":"<div><div>Histopathological images are regarded as the gold standard in cancer diagnosis. Formalin-fixed paraffin-embedded (FFPE) tissues are routinely collected and archived for pathological examination. However, the time-consuming procedures of tissue fixation and embedding render FFPE tissues unsuitable for intraoperative diagnosis, where immediate results are crucial during surgical procedures. In contrast, obtaining a fresh frozen section (FS) takes a very short time. FS samples are widely utilized for intraoperative diagnosis, whereas the diagnostic accuracy of FS is currently limited by the presence of potential histological artifacts. In this paper, we propose a contrastive learning image translation and multiple instance learning network (CoLM) for lung cancer classification. CoLM efficiently translates FS images into FFPE-style images and facilitates whole slide image classification. The entire framework encompasses two crucial stages. In the first stage, we employ a contrastive learning translation network with a dual-attention module (CL-DAM) for image translation. In the second stage, we utilize a hybrid transformer multi-instance learning-based network (HTM) to address the challenge posed by weak labels. We conduct experiments on lung cancer datasets to validate the performance of our proposed approach. The results demonstrate that our method achieve superior classification performance over other state-of-the-art methods, effectively mitigating the impact of blurred FS images. The proposed framework not only elevates the precision of intraoperative diagnosis when employing FS but also provides valuable reference for pathologists through the application of synthetic images.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"100 ","pages":"Article 107097"},"PeriodicalIF":4.9,"publicationDate":"2024-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142553061","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

Cervical vertebral maturation assessment using an innovative artificial intelligence-based imaging analysis system 利用创新型人工智能成像分析系统评估颈椎成熟度

IF 4.9 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biomedical Signal Processing and Control

Pub Date : 2024-10-30 DOI: 10.1016/j.bspc.2024.107088

Hossam Magdy Balaha , Ahmed Alksas , Amine Fattal , Amir A. Sewelam , Wael Aboelmaaty , Khaled Abdel-Ghaffar , Toru Deguchi , Ayman El-Baz

The Cervical Vertebral Maturation (CVM) assessment plays a pivotal role in orthodontic diagnosis and treatment planning by providing insights into skeletal growth and enabling timely interventions. This study introduces an innovative approach to predict CVM stages based on novel imaging markers extracted from X-ray images, which are then correlated with CVM stages. The proposed system comprises the following main steps: (i) initiating with manually delineated cervical vertebrae (i.e., C2, C3, and C4) from the X-ray images; (ii) parcellating the cervical vertebrae based on the Marching level-sets approach to generate five iso-contours for each segmented cervical vertebra; the primary objective of vertebrae segmentation is to extract both local and global imaging markers to accurately grade and classify CVM stages; (iii) extracting first and second-order appearance and morphology imaging markers that describe the shape and appearance of each extracted cervical vertebra; and (iv) employing two-stage classifiers to grade and classify CVM for each patient. The system without data augmentation demonstrated promising results, achieving an accuracy of 95.85%, sensitivity of 88.03%, specificity of 97.20%, and precision of 88.70%. After applying data augmentation techniques, the accuracy improved to 98.89%, with a mean score of 97.20%. To the best of our knowledge, this is the first system to assess the six stages of CVM with such high accuracy. The proposed AI-based system will enhance orthodontic patient care in the USA and worldwide by providing a new non-invasive tool for early CVM assessment.

颈椎发育成熟度（CVM）评估通过深入了解骨骼生长情况和及时干预，在正畸诊断和治疗计划中发挥着关键作用。本研究介绍了一种基于从 X 光图像中提取的新型成像标记预测 CVM 阶段的创新方法，然后将这些标记与 CVM 阶段相关联。拟议的系统包括以下主要步骤：(i) 以人工划定的颈椎（即 C2、C3 和 C4）为起始点、椎体分割的主要目的是提取局部和全局成像标记，以准确分级和划分 CVM 阶段；(iii) 提取描述每个提取的颈椎的形状和外观的一阶和二阶外观和形态成像标记；以及 (iv) 采用两阶段分类器对每个患者的 CVM 进行分级和分类。未使用数据增强技术的系统取得了可喜的成果，准确率达到 95.85%，灵敏度达到 88.03%，特异性达到 97.20%，精确度达到 88.70%。应用数据增强技术后，准确率提高到 98.89%，平均得分 97.20%。据我们所知，这是首个能以如此高的准确率评估 CVM 六个阶段的系统。所提出的基于人工智能的系统将为早期 CVM 评估提供一种新的无创工具，从而提高美国乃至全球的正畸患者护理水平。

{"title":"Cervical vertebral maturation assessment using an innovative artificial intelligence-based imaging analysis system","authors":"Hossam Magdy Balaha , Ahmed Alksas , Amine Fattal , Amir A. Sewelam , Wael Aboelmaaty , Khaled Abdel-Ghaffar , Toru Deguchi , Ayman El-Baz","doi":"10.1016/j.bspc.2024.107088","DOIUrl":"10.1016/j.bspc.2024.107088","url":null,"abstract":"<div><div>The Cervical Vertebral Maturation (CVM) assessment plays a pivotal role in orthodontic diagnosis and treatment planning by providing insights into skeletal growth and enabling timely interventions. This study introduces an innovative approach to predict CVM stages based on novel imaging markers extracted from X-ray images, which are then correlated with CVM stages. The proposed system comprises the following main steps: (i) initiating with manually delineated cervical vertebrae (i.e., C2, C3, and C4) from the X-ray images; (ii) parcellating the cervical vertebrae based on the Marching level-sets approach to generate five iso-contours for each segmented cervical vertebra; the primary objective of vertebrae segmentation is to extract both local and global imaging markers to accurately grade and classify CVM stages; (iii) extracting first and second-order appearance and morphology imaging markers that describe the shape and appearance of each extracted cervical vertebra; and (iv) employing two-stage classifiers to grade and classify CVM for each patient. The system without data augmentation demonstrated promising results, achieving an accuracy of 95.85%, sensitivity of 88.03%, specificity of 97.20%, and precision of 88.70%. After applying data augmentation techniques, the accuracy improved to 98.89%, with a mean score of 97.20%. To the best of our knowledge, this is the first system to assess the six stages of CVM with such high accuracy. The proposed AI-based system will enhance orthodontic patient care in the USA and worldwide by providing a new non-invasive tool for early CVM assessment.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"100 ","pages":"Article 107088"},"PeriodicalIF":4.9,"publicationDate":"2024-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142551985","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