Chinese Journal of Biomedical Engineering

Home

Editorial Board

Download

Quick Search

Office Online

Current Issue

Download Articles

Download

Links

2023 Vol. 42, No. 3 Published: 2023-06-20

Reviews Regular Papers

Regular Papers

	257	Research on the Effect of Visual Presentation Speed on Working Memory Based on EEG Brain Network
		Wang Bixiao, Chen Yao, Li Xin, Wang Shenglin, Huang Liya
		DOI: 10.3969/j.issn.0258-8021.2023.03.001
		With the quickening pace of life, multiple playbacks have been widely used in learning process, and the impacts on the learning cognitive activities have gradually attracted attentions of researches. In order to explore the effect of visual presentation speed of memory tasks on working memory and its mechanism, this paper studied from the view of EEG brain network. An experimental paradigm was designed under the two presentation speeds of fast and slow. EEG data of 18 subjects were collected, power spectrum of each frequency band was calculated, and the frequency band with significant difference (P<0.05) was selected for analysis. The Granger causality method was used to calculate the causal relationship between brain regions in different frequency bands, and a weighted causal brain network was constructed. The three network characteristics of the network, namely, the degree of entry, the degree of exit, and the clustering coefficient, were analyzed. Support vector machines were used to classify the brain networks in the fast and slow state. The results showed that in the fast visual presentation state, the access of the brain network increased, and the node clustering coefficient was further strengthened, and the nodes with significant differences were mainly distributed in the frontal lobe, parietal lobe and occipital lobe, which were significantly higher than those in the slow visual presentation state (P<0.05). The classification accuracy of the brain network in the fast and slow state was up to 90.96%, 90.29% and 86.53% respectively by taking the access, exit and clustering coefficient of each frequency band as characteristics. This study showed that with the acceleration of visual presentation and the further activation of visual processing, the subjects' working memory awareness activities were gradually enhanced, and the leading role of the left hemisphere of the brain in cognitive activities such as language and reasoning were also continuously strengthened. In conclusion, this paper investigated the directed brain network in two states, fast and slow, which provided a new insight for exploring the impact of playback speed on learning cognitive activities, as well as a theoretical basis for learning video designers to set playback speed.
		2023 Vol. 42 (3): 257-265 [Abstract] ( 333 ) HTML (1 KB) PDF (4567 KB) ( 454 )

	266	Study of Binocular Different-Frequency Coding SSVEP-Based Augmented Reality Brain-Computer Interface
		Liu Peishuai, Ke Yufeng, Du Jiale, Ming Dong
		DOI: 10.3969/j.issn.0258-8021.2023.03.002
		In recent years, to improve the flexibility and portability of the BCI system, steady-state visual evoked potential (SSVEP) based brain-computer interface (BCI) and argument reality (AR) technology are combined. However, the performance of AR-BCI based on the traditional SSVEP paradigm is generally low. AR devices can project to both eyes separately. Based on this feature, we proposed a binocular different-frequency coding SSVEP paradigm to stimulate both eyes with different frequencies. This method improved the information content of SSVEP. Fourteen subjects participated in the experiment. Task related component analysis (TRCA) algorithm was used for SSVEP recognition, and the performance of binocular different-frequency coding and binocular same-frequency coding in an AR environment were compared. The spectrum characteristics, signal-to-noise ratio, and spectrum entropy of SSVEP signals under two conditions were analyzed. The classification accuracy of binocular different-frequency coding visual stimulation with different EEG signals time lengths 1, 2, and 3 s could reach 90.9%, 93.9%, and 95.0%, respectively. The classification accuracy of binocular same-frequency coding visual stimulation with different EEG signals time lengths 1, 2, and 3 s could reach 81.1%, 87.8%, and 90.1%, respectively. When the time length was less than or equal to 1 s, the classification accuracy of binocular different-frequency coding visual stimulation was significantly higher than that of binocular same-frequency coding visual stimulation (0.5 s: t (13)=4.562, P<0.01, Cohen's d=1.219; 1 s: t (13)=2.737, P<0.05, Cohen's d=0.732). According to the results of feature analysis, there was no significant difference in the signal-to-noise ratio between the two conditions (t(13)=-1.014, P>0.05, Cohen's d=-0.271), while the spectrum entropy of binocular different-frequency coded signal was significantly higher than that of binocular same-frequency coded signal (t(13)=-2.968, P<0.05, Cohen's d=-0.793). In conclusion, the performance of the AR-BCI system could be improved by the binocular different-frequency coded stimulation.
		2023 Vol. 42 (3): 266-273 [Abstract] ( 230 ) HTML (1 KB) PDF (5173 KB) ( 334 )

	274	Early Diagnosis of MCI and Analysis of Sensitive Brain Regions Based on Multi-Scale Entropy Feature Optimization Algorithm
		Yang Changjie, Li Xin, Hou Yongjie, Wang Yulin, Liu Qinshuang, Su Rui
		DOI: 10.3969/j.issn.0258-8021.2023.03.003
		Mild cognitive impairment (MCI) is an important stage of early diagnosis and timely treatment in the course of Alzheimer's disease. Therefore, early detection and early intervention are urgently needed. Aiming at the problems of early diagnosis of MCI, especially the location of sensitive brain areas in early diagnosis, an optimization algorithm of EEG feature extraction based on multi-scale entropy, namely multi-scale entropy feature optimization algorithm, was proposed. The algorithm was able to mine detailed information to the greatest extent by constructing multi-scale sequences and fully considering the contribution of each sequence. During the experiment, we collected the clinical EEG data of 49 subjects, including 28 in the experimental group (MCI group) and 21 in the normal control group. The entropy values of 16 channel multi-scale entropy feature optimization algorithm in MCI group were lower than those in the control group, and there were significant differences in the prefrontal lobe, anterior temporal lobe and middle temporal lobe brain regions (P<0.01). Only this feature was used as the input feature of the classifier to analyze the three brain regions of prefrontal lobe, anterior temporal lobe and middle temporal lobe, and the recognition rates of the brain region diagnosis test set were 83.33%, 86.67% and 73.33% respectively. Further, the AUC values of the two channels of the anterior temporal lobe with the highest recognition rate were calculated to be 0.753 and 0.733, respectively. The results showed that the entropy feature of the multi-scale entropy feature optimization algorithm could fully reflect the changes of EEG signals and could be used as a potential biomarker for the early diagnosis of MCI. The anterior temporal lobe brain region can provide research support for the sensitive brain region to evaluate the state of brain cognitive function in patients with MCI.
		2023 Vol. 42 (3): 274-280 [Abstract] ( 221 ) HTML (1 KB) PDF (1086 KB) ( 416 )

	281	Dual Dimension Reduction and Channel Attention Gate U-Shaped Network for Pancreatic CT Segmentation
		Ji Jianbing, Chen Shu, Yang Yuanyuan
		DOI: 10.3969/j.issn.0258-8021.2023.03.004
		Segmentation and reconstruction of pancreatic 3D model from CT images is of great significance for assisting diagnosis of the disease. Due to the small proportion of pancreas and the difficulty to distinguish pancreas from the surrounding tissues, existing methods are not accurate and stable enough. Herein we proposed a dual dimension reduction (DDR) and channel attention gate (CAG) U-shaped network. The DDR was added to the coding path to strengthen the effective information extraction in the shallow feature space, and the CAG was embedded in the skip connection to filter redundant feature information from channel level. On the public data set of pancreas segmentation published by NIH (including 82 CT samples), we evaluated the segmentation performance by dice similarity coefficient (DSC), recall (R) and precision (P), and evaluated 3D reconstruction through vertex distance error (VDE). DSC,R and P indexes reached (82.35±5.76)%, (81.07±8.50)%, and (84.04±5.40)% respectively, and VDE decreased to 1.27±0.90, which was better than that obtained from U-Net, Attention-Unet and other methods in the experiment. The experimental results showed that the method we proposed improved the segmentation performance of pancreatic CT images, and the 3D reconstruction model reflected the actual situation of individual pancreas.
		2023 Vol. 42 (3): 281-288 [Abstract] ( 250 ) HTML (1 KB) PDF (6258 KB) ( 438 )

	289	Faster Lung Vessel Segmentation Using Multi-Dimensional Information Fusion
		Chen Huagui, Zhou Lei, Zhao Lian, Cong Zhiyang
		DOI: 10.3969/j.issn.0258-8021.2023.03.005
		The segmentation of vessels in lung CT images plays an important role in the diagnosis and surgical treatment of diseases. In medical image related tasks, deep learning is widely used due to their strong feature representation and discriminative learning abilities. However, deep learning-based methods requires expensive GPUs and large amount of labeled data. In order to achieve better balance between the accuracy and efficiency of vessel segmentation in lung CT images, in this paper, a faster and more effective unsupervised vessel segmentation algorithm based onmulti-dimensional information fusion (MDF) was proposed. The algorithm designed 2D segmentation branches and 3D segmentation branches to make full use of 2D and 3D information, and combined the results of multiple branches in the final segmentation results, which can be quickly and effectively integrated into the traditional unsupervised algorithm. Meanwhile, MDF has strong parallelism capability and can be significantly accelerated on GPU. The proposed MDF based vessel segmentation algorithm was evaluated on the challenging VESSEL12 dataset and CARVE14 dataset comprehensively. The experimental results showed that MDF segmented the vessels in lung CT images with higher accuracy compared with other unsupervised methods. On the CARVE14 dataset, the DSC coefficient of vessels reached 0.716. Moreover, the MDF inference speed was roughly 20 times faster than Frangi algorithms, a multiscale algorithm based on Hession matrix, by GPU parallel optimization. Compared with deep learning-based methods, the segmentation performance of MDF exhibited stronger adaptation ability.
		2023 Vol. 42 (3): 289-300 [Abstract] ( 270 ) HTML (1 KB) PDF (9154 KB) ( 221 )

	301	Multi-Modal Classification Based on Feature Learning of EEG and fNIRS Brain Topographic Map
		He Qun, Xu Xiangyuan, Jiang Guoqian, Shan Wei, Tong Yunjie, Xie Ping
		DOI: 10.3969/j.issn.0258-8021.2023.03.006
		The brain topographic map can be used to monitor the state of brain activity. In order to accurately extract the spatial characteristics of the signals generated by the brain activity of the subjects and effectively improve the classification accuracy, a multi-modal brain topographic map neural network classification algorithm (MBTMNN) was proposed by combining the brain topographic map and convolutional neural network to classify and recognize motor imagery and mental arithmetic. Firstly, EEG and fNIRS signals were preprocessed to extract the energy characteristics of EEG and the concentration characteristics of oxygenated hemoglobin in fNIRS. The colormap of all samples was unified by combining with the position of each electrode to generate brain topographic maps. The EEG and fNIRS signals were simultaneously input into the convolutional neural network and fused in the feature layer to obtain a training model. The six-fold cross validation experiment was conducted on the 2017 Berlin EEG/fNIRS public dataset. The dataset included 29 subjects, with 300 samples each, in the four classification scenarios of left versus right hand motor imagery, mental arithmetic versus resting state, motor imagery versus mental arithmetic versus resting state, and left versus right hand motor imagery versus mental arithmetic versus resting state, the accuracy rates were 82.91%, 94%, 90.34% and 78.18%, respectively, which were higher than those of recently reported with the same dataset and the state of single mode method. The results indicated that the proposed method effectively fused EEG and fNIRS signals to improve the classification accuracy.
		2023 Vol. 42 (3): 301-310 [Abstract] ( 215 ) HTML (1 KB) PDF (7498 KB) ( 235 )

	311	Image-Guided Tele-Operated Robotic Ventricular Puncture System
		Yang Xiaohan, Sun Zhen, Qi Yansong, Wang Junchen
		DOI: 10.3969/j.issn.0258-8021.2023.03.007
		Ventricular puncture and drainage is an important first aid method for craniocerebral injury. However, there are some problems such as time-consuming and labour-intensive craniotomy, and difficult to perform without experiencedsurgeons. In order to simplify the first aid process and reduce the requirements of emergency surgery, an image-guided teleoperation ventricular puncture robot system was developed. The system included a teleoperation puncture actuator system, a robotic arm, a preoperative planning system, and a visual navigation system. The pressure sensor differential layout was used to design the skull puncture actuator based on EtherCAT bus. Variability registration based on B-spline transformation was used for surgical planning and visual navigation on CT and MR fused images. Under the guidance of binocular vision, the 7 degree of freedom redundant robotic arm was used to locate the skull puncture actuator to the starting position according to the preoperative planning path and the specified attitude. The operation was visualized by visual tool library (VTK) in human-in-loop operation mode, and the skull puncture was performed under double feedback of visual and force perception. The puncture experiment on the 3D printed skull model showed that the puncture error of the robot was 0.74 mm. Animal experiments on a beagle dog showed that the puncture error was 1.22 mm, which was comparable to the accuracy of experienced emergency physicians, indicating that the robot held the potential for clinical application.
		2023 Vol. 42 (3): 311-320 [Abstract] ( 177 ) HTML (1 KB) PDF (8066 KB) ( 115 )

	321	Preparation and Characterization of Platelet Membrane Biomimetic Nanocarriers and Transplacental Barrier Transport Efficiency in vitro
		Hu Danhui, Pan Yuxue, Wang Peng, Jiao Zhenna, Wan Guoyun, Wang Haijiao, Sui Junhui, Tang Hongbo, Chen Hongli
		DOI: 10.3969/j.issn.0258-8021.2023.03.008
		This study aims to construct a platelet membrane biomimetic nanocarrier and evaluate the transport efficiency across the placental cell barrier in vitro. The poly(lactic-co-glycolic acid) nanoparticles coated with platelet membrane (P-PLGA NPs) were fabricated by different approach including sonication, coextrusion and covalent connection. The physicochemical characteristics of the P-PLGA NPs were investigated by particle size analyzer, TEM, SDS-PAGE and Elisa, by which the optimal preparation method was determined. Cytotoxicity of the nanocarriers to the cell line BeWo b30 were evaluated by MTT assay. The influence of different endocytic inhibitors on the cellular uptake was examined using flow cytometry. The transwell device was used to investigate the transplacental barrier efficiency of the nanocarriers in vitro. Results showed that the size of P-PLGA NPs prepared by sonication, physical extrusion and covalention was (269.1±32.9), (425.0±36.6), and (823.4±73.1) nm, respectively, which were higher than PLGA NPs, and the value of zeta potential was close to the platelet membrane. The surface membrane proteins were intact and specific protein marker P-selectin still present. The cell viability in all groups had no significant difference compared with the control group (P>0.05). Compared with the control group, the cellular uptake inhibition of NY was 53% in the PLGA NPs group and that of AMR was 45% in P-PLGA NPs group, and the uptake pathway was caveolin-mediated endocytosis and macropinocytosis. In the placental cell barrier model, the PLGA NPs transport efficiency was 6% and 24.3% higher at apical chamber administration concentrations of 5 and 20 μg/mL, respectively, and the P-PLGA NPs transport efficiency was PLGA NPs 13.8% higher at 100 μg/mL. The PLGA NPs transport efficiency was 6% and 24.3% higher than P-PLGA NPs at apical chamber administration concentrations of 5 and 20 μg/mL, respectively, and the P-PLGA NPs transport efficiency was 13.8% higher than PLGA NPs at 100 μg/mL. In conclusion, the platelet membrane was successfully coated on the surface of PLGA NPs. The P-PLGA NPs prepared by sonication had a smaller size and better size distribution. When the nanocarriers' concentration was increased, P-PLGA NPs showed significant increasing transport efficiency in the placental barrier.
		2023 Vol. 42 (3): 321-327 [Abstract] ( 231 ) HTML (1 KB) PDF (4444 KB) ( 226 )

	328	Ferric Chloride and Tannic Acid Treatment Enhance the Comprehensive Properties of Bovine Jugular Vein
		Wang Aili, Zhou Jianye, Wang De, Zhou Qingliang, Wang Yumiao, Huo Meijun, Wang Yibo
		DOI: 10.3969/j.issn.0258-8021.2023.03.009
		Glutaraldehyde (GA) cross-linked bovine jugular vein (BJV) is prone to calcification, which affects its wide clinical application. The addition of tannic acid (TA) treatment in combination with GA cross-linking has achieved excellent anti-calcification effect, however, with a deficiency of poor tissue compliance. As tannins can chelate with metal cations, this study used ferric ions (Fe³⁺) to interact with tannins to improve the comprehensive performance of TA treated BJV. In this study, the addition schemes of Fe³⁺ and reaction conditions such as pH were explored, the mechanical properties were examined by the uniaxial tensile test, the rat subcutaneous implantation model was used to determine the calcification level, and two-point bending method was used to test the flexibility. Vessels obtained from optimized treatment protocols were tested for biocompatibility. The results showed that under pH=8, the addition of Fe³⁺ before TA treatment improved the tissue compliance, and the flexibility was significantly improved compared with that of TA group, meanwhile, the mechanical strength was maintained. Calcium contents after subcutaneous implantation in rats were significantly lower than that in GA group [21d:(1.71±0.41) mg/g vs (38.12±7.40) mg/g, 60d:(2.73±1.13) mg/g vs(124.19±14.22) mg/g, P<0.05], and were equal to that of TA group (P>0.05). Moreover, the biocompatibility of theBJV met the requirements of implantable medical devices. This TA-Fe³⁺ treatment scheme is expected to be a new method for anti-calcification of BJV.
		2023 Vol. 42 (3): 328-335 [Abstract] ( 148 ) HTML (1 KB) PDF (6740 KB) ( 263 )

	336	Design and Characterization of Injectable Drug-Loaded Hydrogels for Myocardial Infarction Therapy
		Liu Yang, Ma Yaxin, Ying Qiuqiu, Zheng Huilin, Zhang Lei
		DOI: 10.3969/j.issn.0258-8021.2023.03.010
		Preventing cardiac fibrosis is the main strategy to improve cardiac function. In this study, the extracellular matrix of myocardium (ECM) and doxycycline composite hydrogels were designed and developed to solve the problems of tissue regeneration and rapid fibrin deposition in myocardial infarction sites. Porcine cardiacs were subjected to SDS, TritonX-100 treatment to remove intracellular material. Then dissolved with pepsin and added fibroin microcapsules containing 54.63~89.27 μg/mg doxycycline. The mixed solution wasspontaneouslyformed into hydrogel at 37℃. The structure of composite hydrogel microcapsules and fibroin microcapsules were observed by scanning electron microscopy, and the feasibility of composite hydrogel treatment was tested by in vitro degradation, sustained drug release and cardiac fibroblast culture experiments. Finally, myocardial infarction rats were treated in vivo (n=9) and infarct size and hemodynamic changes were analyzed to simulate the clinical therapeutic effect. The results showed that the composite hydrogel had a dense pore structure. Compared with the drug-loaded microcapsules, the porous fibroin microcapsules with a diameter of 1~3 μm were evenly distributed in the hydrogel, which not only achieved slow release of the drug, but also significantly prolonged the degradationtime of the hydrogel. In the compound hydrogel, 13% of the total drug load was slowly released for 9 days, and the amount of collagen deposition in fibroblasts was reduced. In vivo treatment studies, the composite water gel inhibited cardiac fibrosis and significantly reduced the myocardial infarction area. After 4 weeks of treatment, LVSP and (dp/dt)_max of myocardial infarction rats increasedto (116.67±3.50) and (2359.24±133.06) mmHg/s compared with (73.52±4.24) and (1 020.96±100.68) mmHg/s in control group. There were significant differences in both of them (P<0.01), which indicated that the pumping ability of the heart was enhanced. In conclusion, the composite hydrogel has good biocompatibility, slow release function, the ability to inhibit fibrosis and enhance cardiac function, It's expected to play a role in clinical treatment.
		2023 Vol. 42 (3): 336-344 [Abstract] ( 172 ) HTML (1 KB) PDF (8549 KB) ( 226 )

Reviews

	345	Research Progress on Potential Brain Stimulation Targets of rTMS for Alleviating Motor Symptoms in Parkinson's Disease
		Li Runze, Yang Shuo, Feng Keke, Wang Alan, Tian Shuxiang, Yin Shaoya, Xu Guizhi
		DOI: 10.3969/j.issn.0258-8021.2023.03.011
		Repetitive transcranial magnetic stimulation (rTMS) is widely used in the clinical treatment of psychiatry and neurology, which can improve the motor symptoms of Parkinson's disease (PD), and is an important approach to realize the precision treatment strategy of PD. The pathological changes in basal ganglia circuit lead to diverse clinical types of PD. To understand the modulatory effects of rTMS in key target areas according to different motor symptoms is significant to improve the therapeutic effect. At the beginning of this paper, the effect of rTMS on PD from the perspective of synaptic plasticity and neural circuits was analyzed. Then, the potential stimulation targets of primary motor cortex (M1), dorsolateral prefrontal cortex (DLPFC), supplementary motor cortex (SMA) and cerebellum were reviewed to discuss the modulatory effects of different stimulation parameters in specific brain regions and the effects of different stimulation target areas on different motor symptoms. At last, the main problems in the current research were pointed out, and the future research trends were discussed from the aspects of multi-target stimulation, optimal intervention period, and combination with brain imaging technology.
		2023 Vol. 42 (3): 345-352 [Abstract] ( 248 ) HTML (1 KB) PDF (1238 KB) ( 479 )

	353	Research Progress of Radiomics and Deep Learning in Pancreatic Cancer
		Chen Can, Bao Jiayi, Hu Su
		DOI: 10.3969/j.issn.0258-8021.2023.03.012
		Imaging examinations are used for preoperative diagnosis and evaluation of pancreatic cancer, which play great role in clinical decision-making and prognosis prediction. Radiomics and deep learning provide new means for accurate diagnosis and treatment of pancreatic cancer. In this paper, the developments of radiomics and deep learning in pancreatic cancer were reviewed. The paper reviewed the application of radiomics and deep learning in predicting malignant potential of precursor lesions, diagnosis and differentiating diagnosis, therapeutic effect evaluation, prognosis prediction, and gene expression status. Radiomics and deep learning may provide comprehensive information for clinical decision-making and provide help for precise diagnosis and treatment of pancreatic cancer, with standardized quality control and multi-center database establishment.
		2023 Vol. 42 (3): 353-359 [Abstract] ( 285 ) HTML (1 KB) PDF (822 KB) ( 978 )

	360	Cryoablation Combined with Immunotherapy: Maximizing Anti-Tumor Effects
		Huang Ruotong, Lui Baolin
		DOI: 10.3969/j.issn.0258-8021.2023.03.013
		In recent years, energy therapy-based tumor ablation techniques have become more widely used. Among them, the cryoablation technique can effectively destroy tumors by rapid cooling of the probe, and the tumor mass left in situ would release a lot of antigens to activate immune cells, inducing the generation of anti-tumor immune responses. However, the immune response triggered by cryoablation is subjected to its conditional setting, and the intensity and sustainability of the immune response are insufficient to suppress tumor recurrence and the growth of metastases. Hence, this review focuses on the mechanisms of cryoablation for tumors, including direct cell injury, vascular injury, and immunomodulation, in which immunomodulation has a significant impact on the favorable prognosis of tumor therapy. Cryoablation factors affecting immune efficacy were further introduced to discuss the optimal conditions for generating immune response after treatment. The research progress of cryoablation combined with immunotherapy is summarized to propose new strategies for tumor treatment.
		2023 Vol. 42 (3): 360-369 [Abstract] ( 236 ) HTML (1 KB) PDF (2641 KB) ( 569 )

	370	Research on the Applications of Electrochemical Sensor Modified with Layer-by-Layer Self-Assembly Films
		Shi Gaofan, Lin Xiangde, Liu Huajie, Zhang Mengmeng, Xia Pengpeng, Liu Sisi, Chen Yuzhu, Zeng Dongdong
		DOI: 10.3969/j.issn.0258-8021.2023.03.014
		Electrochemical sensors are widely used in the detection of various biochemical substances due to their excellent sensitivity, detection limit, selectivity and response rate. The electrode acts as a sensitive element of electrochemical sensors, and a reasonable surface modification coating is particularly important to improve the detection accuracy and stability of the sensor. The layer-by-layer self-assembly technology is widely used in the field of electrochemical sensors, because it is able to construct three-dimensional material system with nanometer precision controllable parameters by alternately depositing interacting substances on the substrate. In this paper,we summarized assembly materials (including polyelectrolytes and nanoparticles, etc.)the film driven force (including electrostatic interaction, covalent bond, hydrogen bond, etc.), performance optimization and technical improvement of the layer-by-layer self-assembly technologies were summarized. In addition, electrochemical sensors modified by layer-by-layer self-assembly films of nanomaterials and their applications in biomarker detection, air monitoring, metal ion detection were introduced.
		2023 Vol. 42 (3): 370-384 [Abstract] ( 167 ) HTML (1 KB) PDF (3869 KB) ( 476 )

Copyright © Editorial Board of Chinese Journal of Biomedical Engineering
Supported by:Beijing Magtech