首页期刊导航|Biomedical signal processing and control
期刊信息/Journal information
Biomedical signal processing and control
Elsevier
Biomedical signal processing and control

Elsevier

季刊

1746-8094

Biomedical signal processing and control/Journal Biomedical signal processing and controlEISCIISTPAHCI
正式出版
收录年代

    An improved version of firebug swarm optimization algorithm for optimizing Alex/ELM network kidney stone detection

    Ding H.Huang Q.Razmjooy N.
    1.1-1.11页
    查看更多>>摘要：© 2024 Elsevier LtdThe use of CT scan to diagnose kidney stones is among the most accurate ways to confirm the presence of kidney stones in patients. The scan takes photographs inside the body using a computer and an X-ray. The present study proposes a new automatic methodology using an integrated Alexnet and ELM (Extreme Learning Machine) network to deliver more useful outcomes of detection for kidney stone. Afterward, the network is optimized on the basis of a newly improved version of firebug swarm optimization algorithm. The designed network is applied to the “CT Kidney Dataset”, and its outcomes are then verified by some different advanced procedures. The final results indicated that the proposed approach has better performance than the other methods.
      原文链接:
    • NETL
    • NSTL
    • Elsevier

    Point out the mistakes: An HMM-based anomaly detection algorithm for sleep stage classification

    Wang Z.Liu H.Cai Y.Li H....
    1.1-1.9页
    查看更多>>摘要：© 2024 Elsevier LtdAccurate sleep stage scoring is essential for diagnosing sleep disorders. Current automated sleep staging methods often exhibit staging errors, which can be interpreted as anomalies. Detecting these anomalies is crucial for improving staging accuracy. Most existing approaches modify staging based on predefined conditions but lack effective methods for localizing and identifying anomalies. In this study, we propose an anomaly detection method utilizing the Hidden Markov Model (HMM), a time-series modeling technique, to detect anomalies in sleep staging results. Evaluating our approach with four classical models as pre-classifiers, we achieve anomaly detection precisions of 0.760, 0.577, 0.631, and 0.613. Assuming that all detected anomalies are corrected, the pseudo-accuracies improve to 0.964, 0.929, 0.950, and 0.929, respectively. Our results indicate that the proposed method significantly enhances stage recognition accuracy, especially for stage N1, which is critical for diagnosing sleep-related disorders. Notably, approximately 28.6% of epochs require reinterpretation by sleep technicians to achieve these improvements.
      原文链接:
    • NETL
    • NSTL
    • Elsevier

    Thin vessel segmentation in fundus images using attention UNet and modified Frangi filtering

    Varma, AnumehaAgrawal, Monika
    1.1-1.9页
    查看更多>>摘要：Thin vessel segmentation is an active research problem, with an emphasis on finding a universal approach for different types of fundus datasets. Enhancement of the thin vessels is the first and foremost task for proper segmentation, which is proposed to be done with the total variation (TV) decomposition method with layer- selective enhancement and illumination correction. The vessel segmentation task is carried out on two fronts. For thin vessels, we propose the attention UNet backbone, and for thick vessels, the modified Frangi method is used. The method is trained and tested on HRF, CHASE_DB1, and DRIVE datasets with accuracy of 96.73%, 96.38%, and 94.97%, and sensitivity of 92.66%, 81.05%, and 87.56%, respectively. The method was cross- validated on the STARE dataset with an accuracy of 96.30% and a sensitivity of 93.69%. The sensitivity performance surpasses the state of the art.
      原文链接:
    • NETL
    • NSTL
    • Elsevier

    Elevated correlations in cardiac–neural dynamics: An impact of mantra meditation on stress alleviation

    Singh S.Gupta K.V.Behera L.Bhushan B....
    1.1-1.16页
    查看更多>>摘要：© 2024 Elsevier LtdMantra meditation is a widely practiced technique with a demonstrated ability to promote relaxation and alleviate stress. This study aims to evaluate the impact of regular mantra meditation practice on electroencephalogram (EEG) patterns and heart rate variability (HRV) in young adults grappling with increased stress levels. Furthermore, the study explored possible correlations between these two physiological markers. Thirty-eight participants practiced mantra meditation for eight weeks, and their EEG and HRV were recorded before and after. The pre-post analysis evaluated brain activity and autonomic nervous system regulation alterations by examining EEG and HRV in the time, frequency, and nonlinear domains. Data analysis revealed significant correlations between EEG patterns and HRV indices following mantra meditation, suggesting a potential interconnection between neural and cardiovascular regulation. A notable increase in Root Mean Square of Successive Differences (RMSSD) and Standard Deviation of NN intervals (SDNN) within HRV, along with a decrease in Heart Rate (HR) and Stress Index (SI) and an increase in EEG alpha and theta power, accompanied by a reduction in their ratio, was observed. Three pivotal correlations consistently increased during eight weeks: Alpha/Theta Ratio (ATR) with Low-Frequency/High-Frequency Ratio (LHR), ATR with SD2/SD1 Ratio (SDR), and Relaxation Index (RI) with Stress Index (SI). After eight weeks, these correlations attained statistical significance, optimizing the balance between calmness and vigilance, boosting parasympathetic activity, and reducing chaos between the parasympathetic and sympathetic systems. The study elucidates the scientific mechanisms of mantra meditation, enriching research on its stress-reducing effects and emphasizing mental and physical well-being.
      原文链接:
    • NETL
    • NSTL
    • Elsevier

    EATNet: An extensive attention-based approach for cervical precancerous lesions diagnosis in histopathological images

    Xu J.Shi L.Gao Y.Zhang Y....
    1.1-1.13页
    查看更多>>摘要：© 2024Grading of cervical precancerous lesions is an important prerequisite for determining the treatment plan for precancerous lesions. However, on account of the huge scale of whole slide histopathological images but the small area of interest, the lack of pixel-level annotation data, and the subjectivity of lesion diagnosis without definite quantified standards, which lead to the difficulty of lesion classification. Most existing methodologies split high-resolution images into patches and employ patch-based local feature representations to deliver image-level decisions, resulting in the destruction of the contextual information and the weakened ability to learn clinically relevant representations. To overcome these challenges, this study proposes an Extensive ATtention Network (EATNet) for diagnosing cervical precancerous lesions in histopathological images. EATNet extends the bag-of-words strategy by splitting a whole slide histopathological image into several bags and instances to end-to-end learn representations from gigapixels. The instance-level and bag-level attention blocks are designed to encode the abundant global dependencies, in order to produce discriminative WSI descriptors with only slide-level labels. Experiments are conducted on two public cervical and endometrial datasets, which demonstrate superior performance over prevalent methods with AUC of 92%–94%.
      原文链接:
    • NETL
    • NSTL
    • Elsevier

    High-frequency SSVEP-BCI system for detecting intermodulation frequency components using task-discriminant component analysis

    Cui H.Li M.Chen X.Ma X....
    1.1-1.9页
    查看更多>>摘要：© 2024 Elsevier LtdRecently, steady-state visual evoked potential (SSVEP)-based brain-computer interface (BCI) has significantly progressed and is moving from the laboratory to practical application. However, the system performance and comfort of SSVEP-BCIs still need to be improved. In this study, five flicker frequencies (i.e., 30–34 Hz with an interval of 1 Hz) and eight scaling frequencies (i.e., 0.4–1.8 Hz with an interval of 0.2 Hz) were adopted to jointly encode forty visual stimulus targets using evoked intermodulation (IM) frequency components. Both luminance and shape changes are implemented by sinusoidal sampling stimulus coding methods. High-frequency flicker frequencies and green visual stimuli were chosen to improve the comfort of the proposed system. An extended version of a training algorithm named task-discriminant component analysis (TDCA) was proposed to detect the IM components of SSVEP signals. The average recognition accuracy of eleven subjects is 96.82 ± 0.01 % in the offline experiments for a data length of 5 s. Online validation experiments was constructed from the optimized parameters of offline analysis, and the average accuracy and ITR were 94.37 ± 1.17 % and 113.47 ± 2.60 bits/min, respectively. Furthermore, ten subjects who participated in the validation part also completed the online free-spell task successfully. These results showed that it is feasible to expand the number of stimulus targets by using IM frequency components of SSVEP signals for target coding, and that the system performance is superior.
      原文链接:
    • NETL
    • NSTL
    • Elsevier

    A pyramid convolutional mixer for cervical pap-smear image classification tasks

    Yang T.Hu H.Qing M.Huang Q....
    1.1-1.13页
    查看更多>>摘要：© 2024 Elsevier LtdConvolutional Neural Networks (CNNs) have exhibited considerable success in the realm of cervical cytopathology image classification, owing to their efficient design. We find that existing CNN-based cervical cytopathology classification methods fail to fully exploit the cell morphology and nucleus information. To address the above problems, we propose an efficient network called Pyramid Convolutional Mixer. We capture multi-scale subtle morphology features at the cellular level and convey nuclear neighborhood spatial information by integrating convolutional operations within the transformer structure. PCMixer contains two key modules, i.e. pyramid morphology module (PMM) and nuclear spatial mixing block (NSMB) to retrieve cervical cytopathology information. PMM is characterized by a multi-scale pyramid architecture employing a convolutional layer and a local encoder to generate local morphology information at each scale. In addition, NSMB operates on the input patches to separate the mixing of spatial and channel dimensions to encode nuclear neighborhood spatial information. We intend to unveil a more intricate cervical cytopathology dataset: Cervical Cytopathology Image Dataset (CCID). We achieve a classification accuracy of 89.62% along with precision, recall and F1 score of 82.76%, 85.97% and 84.15% respectively on the CCID dataset. Also, we use cervical cytopathology images from the publicly available SIPaKMeD dataset. We obtain 96.21%, 95.70% 95.60% and 95.30% respectively for the four metrics. Through comprehensive experiments conducted on two real-world datasets, our proposed model demonstrates superior performance compared to state-of-the-art cervical cytopathology classification models. The results demonstrate that our method can significantly assist cytopathologists in appropriately evaluating cervical smears.
      原文链接:
    • NETL
    • NSTL
    • Elsevier

    SAFE-Net: Shape-aware and feature enhancement network for polyp segmentation

    Yu J.Qi L.
    1.1-1.13页
    查看更多>>摘要：© 2024 Elsevier LtdPrecise identification of colon polyps, which are the principal precursors of colorectal cancer, is crucial for accurate diagnosis and treatment. Current segmentation methods possess unique strengths and limitations, given the colon polyp's varying size, shape, color, and unclear edge with surrounding mucosa. These challenges result in the absence of a universally effective segmentation method. Addressing the aforementioned issues, this paper proposes a novel shape-aware and feature enhancement network (SAFE-Net) for polyp segmentation. SAFE-Net contains three innovative modules: the shape-aware module (SAM), the feature enhancement module (FEM), and the multi-modality attention module (MMAM). SAM leverages the disparities in low-level features within the backbone network to effectively filter out noise from colonoscopy image, thereby extracting distinct detailed features, including shape, texture, and edge, ultimately enhancing the precision of polyp segmentation. FEM enhances and enriches the semantic information of the backbone network's features, enables the capture of polyps varying in shape and size, and selects and refines relevant features. MMAM utilizes high-level feature prediction as a guide map to achieve attention on the background, foreground, and boundary, making the network focus more on suspicious and complex polyp regions. This paper conducts experiments on five colonoscopy image datasets: Kvasir-SEG, CVC-ClinicDB, CVC-ColonDB, CVC-T, and ETIS. Comparative analyses are conducted between the proposed model and existing models. The experimental results confirm that the proposed model exhibits superior segmentation capabilities and provides a reliable foundation for subsequent diagnostic procedures.
      原文链接:
    • NETL
    • NSTL
    • Elsevier

    An extended variational autoencoder for cross-subject electromyograph gesture recognition

    Zhang Z.Ming Y.Shen Q.Wang Y....
    1.1-1.13页
    查看更多>>摘要：© 2024 Elsevier LtdSurface electromyographic hand gesture recognition has gained significant attention in recent years, especially within the field of human–computer interfaces. However, cross-subject tasks remain challenging due to inherent individual differences. To address this, a novel approach for hand gesture recognition is proposed that leverages a subject-generalized variational autoencoder. This approach involves an extended variational autoencoder designed to disentangle input data into three distinct feature-specific representations. The primary classifier within the variational autoencoder focuses on gesture recognition, while two auxiliary classifiers work together to extract subject-specific and gesture-specific features. The gesture-specific features capture generalized characteristics applicable across all subjects, enabling direct application to new subjects. To enhance accuracy and stability, a competitive voting strategy is implemented. The effectiveness of the proposed method was evaluated using a dataset comprising six representative gestures performed by eight subjects. Comparative analysis with baseline models shows that our approach outperforms others, demonstrating superior generalization with an average accuracy of 90.52% in cross-subject validation.
      原文链接:
    • NETL
    • NSTL
    • Elsevier

    Structure-aware single-source generalization with pixel-level disentanglement for joint optic disc and cup segmentation

    Jiang J.-X.Li Y.Wang Z.
    1.1-1.9页
    查看更多>>摘要：© 2024 Elsevier LtdDeploying deep segmentation models in new medical centers poses a significant challenge due to statistical disparities between source and unknown domains. Recent advancements in domain generalization (DG) have shown improved generalization performance by leveraging disentanglement techniques on domain-specific and domain-invariant features. However, existing DG methods face challenges in achieving optimal feature segregation. To address this, we introduce a pixel-level contrastive single domain generalization (PCSDG) framework and a structure-aware brightness augmentation (SABA) technique for joint optic disc and cup segmentation. First, a disentanglement module captures content and style-related maps, pixel-wise multiplied with the original image to generate saliency-based attention maps, resulting in distinct structure and style representations. Second, using a contrastive loss in the latent space enhances segregation. Finally, SABA introduces random brightness variations, preserving anatomical information and diversifying sample styles. Experimental validation on two public fundus image datasets with two source domains and five target domains demonstrates the superior performance of PCSDG and SABA across diverse domains when compared to state-of-the-art methods. Our code and models are made public at: https://github.com/HopkinsKwong/PCSDG.
      原文链接:
    • NETL
    • NSTL
    • Elsevier