Chinese Journal of Biomedical Engineering

Home

Editorial Board

Download

Quick Search

Office Online

Current Issue

Download Articles

Download

Links

2022 Vol. 41, No. 2 Published: 2022-04-20

Reviews Communications Regular Papers

Regular Papers

	129	Individual-Level Assessment in Patients with Disorders of Consciousness under Passive Auditory ERP Paradigm
		Wang Xiaoyu, Yang Yi, Li Fan, Chen Xueling, Gao Hanbing, Ma Zhaonan, He Jianghong, Cong Fengyu
		DOI: 10.3969/j.issn.0258-8021.2022.02.001
		In recent years, there have been increased efforts to assess the levels of consciousness in patients with disorders of consciousness (DoC) using neuroimaging techniques, aiming to improve diagnosis and identification beyond current subjective behavioral assessments that suffer from high misdiagnosis rates. Previous evidence suggests that N1 and mismatch negativity (MMN) components elicited by passive auditory event-related potentials (ERPs) paradigms are critical neurophysiological markers of DoC. However, as such evidence is limited to group-level analysis, the extent to which they enable residual consciousness detection at the individual-level is unclear. Considering the characteristics of N1 and MMN components, we proposed a deep learning algorithm for the individual assessment of patients with DoC under a passive auditory ERPs paradigm. The algorithm proposed a data augmentation strategy, which randomly fused single-trials elicited by different types of stimuli in the spatial domain to form fusion samples, and a deep learning classifier, known as EEGNET, to achieve automatic feature extraction and classification. The proposed method was evaluated in a three-class classification task (38 healthy controls, 40 minimally conscious state, and 54 vegetative state patients) using a single-trial dataset including 132 subjects. Statistical results showed that the proposed data augmentation method significantly improved the classification performance in the current task, and it achieved the highest 75.14% mean classification accuracies in sample level as well as 83.00% mean classification accuracies, 83.79% precision rate, and 84.02% recall rate in subject level when the number of single-subject samples was augmented to 1000. In conclusion, the proposed method could overcome the drawbacks of poor assessment performance in the conventional individual-level assessment methods, providing a new strategy for individual-level assessment in patients with DoC.
		2022 Vol. 41 (2): 129-139 [Abstract] ( 350 ) HTML (1 KB) PDF (5779 KB) ( 525 )

	140	Time-Varying Copula Mutual Information for Intermuscular Coupling Analysis
		Wang Hongan, She Qingshan, Ma Yuliang, Kong Wanzeng, Tian Yuping
		DOI: 10.3969/j.issn.0258-8021.2022.02.002
		To measure the relationship between different muscles accurately and reasonably under the regulation of central nervous system is a challenging research topic. Based on the time-varying copula function and combining with the entropy theory, a time-varying copula mutual information (MI) estimation method was proposed in this paper and applied it to the coupling analysis of sEMG signals of biceps brachii (BB) and triceps brachii (TB) in the characteristic frequency bands (theta, alpha and beta) during wrist flexion (WF) and wrist extension (WE) movement of 10 subjects. Meanwhile, the method was compared with the static copula function to verify its effectiveness. The data we used were derived from Ninapro DB4. Experimental results show that compared to the static copula function, the time-varying copula function has a better fitting degree for the intermuscular dependent structure. There was a significant frequency band difference in the intermuscular coupling strength described by the time-varying copula MI (P<0.05), which was specifically expressed as: the higher the frequency band, the lower the intermuscular coupling strength (WF: 0.075 7~0.214 7 bit. WE: 0.078 0~0.237 3 bit), while the static copula MI incorrectly underestimates the intermuscular coupling strength. In conclusion, the time-varying copula MI provided an advanced theoretical guidance method for intermuscular coupling analysis and showed a very broad application prospect.
		2022 Vol. 41 (2): 140-150 [Abstract] ( 203 ) HTML (1 KB) PDF (12508 KB) ( 168 )

	151	Investigating ERP Brain Network for Alcohol-Dependent Patients Based on Phase-AmplitudeCoupling
		Liu Xingping, Wang Suogang
		DOI: 10.3969/j.issn.0258-8021.2022.02.003
		In this study, we explored the application of complex network methods based on the phase-amplitude coupling of event-related potential brain network mechanisms for alcohol-dependent patients. The open-sourced event-related potentials generated by the delayed matching sample paradigm in 76 alcohol-dependent patients and 45 healthy controls were studied. After pre-processing the data, modulation index method was used to calculate the intensity of theta-gamma and alpha-gamma phase-amplitude coupling (TGC, AGC) among all paired channels to construct causal brain network, and then we measured the parameters of brain network graph theory when the network density was 0.3. In addition, the wavelet packet energy of theta, alpha, beta, gamma rhythms and fatigue factors were also calculated. The data was expressed in quartiles of 50% (25%, 75%). The results showed that there was a significant difference in the coupling intensity between the two groups with only non-target stimulation trials. The TGC and AGC between some paired channels of alcohol-dependent patients were significantly decreased (respectively 15.29%, 1.29%, all P<0.05), and the average coupling intensity was significantly decreased, especially for TGC [TGC: 0.014(0.011, 0.018) vs 0.012(0.010, 0.014), P=0.002; AGC: 0.012(0.010, 0.014) vs 0.011(0.010, 0.012), P=0.005]. In the TGC network, the characteristic path length [84.16(60.96, 110.33) vs 104.24(86.93, 118.98), P=0.005] and diameter [222.40(154.78, 254.39) vs 253.39(207.82, 307.99), P=0.003] of alcohol-dependent patients were significantly increased, and the average clustering coefficient [0.013(0.010, 0.019) vs 0.009(0.008, 0.012), P<0.001], global efficiency [0.014(0.011, 0.016) vs 0.011(0.010, 0.013), P<0.001], average local efficiency [0.017(0.013, 0.022) vs 0.013(0.011, 0.017), P<0.001] and transitivity [0.011(0.008, 0.016) vs 0.008(0.007, 0.010), P<0.001] were significantly decreased. The results of AGC network were like that of the TGC network, but the diameter difference was not statistically significant. The wavelet packet energy of theta, alpha, beta rhythms in most electrodes of alcohol-dependent patients were significantly decreased (respectively 95.08%、100%、50.82%, all P<0.05), also the fatigue factors in some electrodes were significantly decreased (45.90%, all P<0.05). When making judgement in non-target stimulation for alcohol-dependent patients, the cortical excitability increased, the inhibition control ability of brain decreased, the function separation and integration ability of TGC and AGC brain network decreased, and the topological organization of brain network was disordered.
		2022 Vol. 41 (2): 151-158 [Abstract] ( 225 ) HTML (1 KB) PDF (6288 KB) ( 272 )

	159	Analysis of Human Body Balance Characteristics in Complex Network Based on Transfer Entropy
		Wang Zheyuan, Luo Zhizeng, Qiu Shenchen
		DOI: 10.3969/j.issn.0258-8021.2022.02.004
		In order to overcome the subjective defects of medical evaluation methods for human balance ability, an algorithm based on phase synchronization screening data and transfer entropy brain network was proposed. A total of 1 200 minutes of EEG balance data were collected from 20 subjects under 4 paradigms of occlusion of proprioception and vision. According to the physiological mechanism of human balance, the effective data segment of the balance adjustment process was screened through the phase synchronization. After data screening, 859 valid data segments were obtained, each with a length of 50~2 000 ms, and the brain function network model based on the transfer entropy was constructed from the screening results. A kind of characteristics was defined that can reflect the reception of human balance information and cooperative processing of motion perception. The 10-fold cross-validation in the same batch of experimental data showed that the new feature improved the classification accuracy of the 4 paradigms to 73.63%, which was higher than that of other brain functional network features (56.23%) and center of pressure (COP) features (67.90%). According to the analysis of synchronization, it is concluded that the visual system plays a major role in the adjustment of human body balance. The application of the new balance features has significantly improved the accuracy of the classification.
		2022 Vol. 41 (2): 159-166 [Abstract] ( 180 ) HTML (1 KB) PDF (3609 KB) ( 359 )

	167	Joint Optic Cup and Disc Segmentation Using Convolutional Neural Network with Receptive Field Module
		Yu Shuyang, Yuan Xin, Zheng Xiujuan
		DOI: 10.3969/j.issn.0258-8021.2022.02.005
		Glaucoma is the world's largest irreversible blindness eye disease. Early diagnosis and timely treatment are effective measures to prevent blindness caused by glaucoma. The cup to disc ratio in fundus images is an important index of early screening and clinical diagnosis of glaucoma. Therefore, accurate segmentation of the optic cup and disc is the key to accurately calculate the cup to disc ratio and improve the computer-aided diagnosis technology of glaucoma. To solve this problem, this paper firstly performed a polar coordinate transformation preprocessing on the fundus image, and then proposed a convolutional neural network Seg-RFNet that integrated the receptive field module to achieve joint segmentation of the optic cup and disc. Seg-RFNet was based on SegNet framework and used ResNet50 as the coding layer to enhance the feature extraction ability of the image, and the coding layer was branched to obtain more deep semantic information. At the same time, the receptive field module was added between the coding layer and decoding layer, which was able to simulate the human visual system, increasing the receptive field and enhance the response of useful features. The 800 fundus images from REFUGE that is a dataset published in MICCAI 2018, were used to verify the performance of the proposed method compared with other methods. The results showed that the Jaccard Similarity (higher is better) of the optic cup and disk segmentation was 0.9515 and 0.8720, and the F score (higher is better) was 0.9749 and 0.9301, respectively. Compared with the commonly used U-Net, SegNet and other networks, Seg-RFNet showed better joint segmentation accuracy of the optic cup and disc and provided an accurate segmentation basis for calculating the cup to disc ratio.
		2022 Vol. 41 (2): 167-176 [Abstract] ( 186 ) HTML (1 KB) PDF (7569 KB) ( 621 )

	177	Classification of Breast Cancer Gene Data Based on gcForest
		Qin Xiwen, Wang Rui, Zhang Siqi
		DOI: 10.3969/j.issn.0258-8021.2022.02.006
		The classification of breast cancer gene data is of great importance in clinical medicine. Aiming at the characteristics of complex structure, high-dimensional and small samples of gene data, this paper proposes a gene data classification method based on the max-relevance and min-conditional redundancy ( mRMCR ) and multi-grained cascade forest ( gcForest ). A total of 98 data were selected from the breast cancer gene expression data set of theBroad Gene Research Institute, and each sample contained 1 213 characteristic genes. Firstly, the data are standardized, then the feature subsets are selected by using the max-relevance and min-conditional redundancy , and finally the feature subsets are classified by the gcForest. Taking random forest, support vector machine and BP neural network as comparison methods, the results show that the best classification accuracy of the proposed combination method of mRMCR and gcForest is 93.78%, which is obviously better than other methods. This method can effectively improve the classification accuracy of breast cancer gene data, and has important theoretical significance and practical value for breast cancer classification based on gene data.
		2022 Vol. 41 (2): 177-185 [Abstract] ( 184 ) HTML (1 KB) PDF (2107 KB) ( 257 )

	186	Prediction of Breast Cancer Neoadjuvant Chemotherapy Based on Longitudinal Time DepthNetwork Fusion
		Xue Tailong, Fan Ming, Chen Shujun, Li Lihua
		DOI: 10.3969/j.issn.0258-8021.2022.02.007
		Neoadjuvant chemotherapy can improve the cure rate of breast cancer, but it is not effective for all patients. Accurate prediction of chemotherapy efficacy can provide reference for physicians to formulate treatment protocols. This study used deep learning to integrate the image characteristics of longitudinal time dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) to predict the efficacy of neoadjuvant chemotherapy. We analyzed 164 DCE-MRI images of patients who underwent neoadjuvant chemotherapy for breast cancer, and selected the maximum tumor diameter and two upper and lower slices from each patient's image data set to expand the data to 442 cases that were randomly divided into 312 cases in the training set and 130 cases in the test set. DCE-MRI images had 6 sequences in total. Segmented the breast area of each sequence and removed the skin and chest cavity. Using deep learning model,the efficacy of neoadjuvant chemotherapy was predicted based on the images before chemotherapy, after 2 courses of chemotherapy and both of them, respectively. We drew the ROC curve of the prediction results and calculated the area under the curve (AUC) to evaluate the classification performance of the model. The best AUC of deep learning model for predicting the efficacy of the images before chemotherapy and the images after two courses of chemotherapy was 0.775 and 0.808 respectively, and the best AUC for predicting the efficacy of the fusion of images before chemotherapy and images after 2 courses of chemotherapy was 0.863, which was better than using the images before chemotherapy. The experimental results showed that compared with the existing approach of using the images before chemotherapy, using the fusion of longitudinal time images could improve the prediction performance of neoadjuvant chemotherapy.
		2022 Vol. 41 (2): 186-194 [Abstract] ( 259 ) HTML (1 KB) PDF (7317 KB) ( 198 )

	195	Research and Application of Pressure Swirl Spray Technique for Blood Smear Staining
		Wang Guowei, Li Wangxin, Mei Qian, Dong Wenfei
		DOI: 10.3969/j.ssn.0258-8021.2022.02.008
		To solve the problems of uneven staining and low efficiency of manual staining blood smear, a method of blood smear spraying based on pressure swirl atomizing nozzle was proposed. Numerical simulation and experiments were performed to analyze the atomization process of Wright-Giemsa stain and investigate the best spray parameters. Firstly, ANSYS software was used to numerically simulate the flow state of the staining solution in the nozzle. Then, the external atomization field test platform of the nozzle was built using a spray laser particle size analyzer and camera. When the external inlet pressure of the nozzle was 0.10, 0.12, 0.14, 0.16, 0.18, 0.20, 0.22 and 0.24 MPa, the spray angle of the staining solution and the atomization particle size parameters (D_v10,D_v90 and droplet volume median diameter D_v50) of the measuring point in the atomization area from 10 mm to 50 mm directly below the nozzle were measured respectively. Afterwards, the distribution span of particle size S was calculated. In addition, blood smear spray staining experiment was carried out to compare the background grayscale parameters of the cell images after drip staining and spray staining. The simulation results showed that staining solution forms a high-speed liquid film at the outlet of the pressure swirl atomizing nozzle, which provided a theoretical basis for the formation of atomization field. The experimental results demonstrated that the best atomization quality was achieved when the nozzle inlet pressure was 0.20 MPa and the spray interface was located 35 mm below the nozzle. The spray angle of the nozzle was 76.13°, the droplet volume median diameter was 75 μm, and the distribution span of particle size was 1.72. Under this condition, the gray parameters of cell image background after spray staining and drip staining were significantly different (P＜0.01), and the gray of cell image background after spray staining was significantly reduced. In conclusion, the pressure swirl atomizing nozzle can be used in blood smear staining to improve the staining efficiency and obtain the uniform staining, which provides a new solution for blood smear staining.
		2022 Vol. 41 (2): 195-203 [Abstract] ( 183 ) HTML (1 KB) PDF (4875 KB) ( 168 )

Reviews

	204	Advances in Multi-Modal Brain-Computer Interface Combined with Steady-State Visual Evoked Potential
		Chi Xinyi, Cui Hongyan, Chen Xiaogang
		DOI: 10.3969/j.issn.0258-8021.2022.02.009
		Brain-computer interface (BCI) is one of the most active research directions in the field of neural engineering, which can establish a communication and control pathway between the brain and the external environment independent of peripheral nerves or muscles and is helpful to restore the self-care ability of people with movement disorders. Steady-state visual evoked potential BCI attracts great attention for the high information transfer rate and less requirements of training. The existing non-invasive high speed BCI is mainly from or based on SSVEP. In recent years, multi-modal BCI integrateing SSVEP and other input signals has become a new trend in BCI research for further improvement of BCI performance. This paper reviewed advances of multi-modal BCI combined with SSVEP from several aspects, including the type of input signals, fusion of experimental paradigm and signal, aiming to help readers understand the research trends in this field and inspire the design and implementation of high communication rate BCI system. Meanwhile, existing problems and possible development trends in the future were discussed to promote the development of multi-modal BCI combined with SSVEP.
		2022 Vol. 41 (2): 204-213 [Abstract] ( 220 ) HTML (1 KB) PDF (1245 KB) ( 791 )

	214	Advance of Transcranial Electrical Stimulation for the Improvement of Motor Performance
		Zhang Na, Liu Hui, Miao Yu, Qi Fengxue
		DOI: 10.3969/j.issn.0258-8021.2022.02.010
		Transcranial electrical stimulation (TES) includes transcranial direct current stimulation, transcranial alternating current stimulation and transcranial random noise stimulation. It is a non-invasive brain stimulation technique using electrodes of different sizes over specific brain regions to modulate cortical neural activity and/or excitability by specific patterns of low-intensity electrical current, thereby strengthen the connections between brain, nerve, and muscle, and improve motor performance. At present, TES technology is going to be utilized to the investigations of sports scientific research. This study first described the neural mechanism of TES over the cerebral cortex, and reviewed the research progress in improving human motor performance of TES over the last two decades in terms of body balance ability, endurance performance muscle fatigue, muscle strength and motor learning. We also reviewed the effects of TES on brain networks functionally connection and discussed the significance of this field in improving motor performance. Finally, we proposed the research perspectives and directions of the TES application in the improvement of motor performance.
		2022 Vol. 41 (2): 214-223 [Abstract] ( 213 ) HTML (1 KB) PDF (1243 KB) ( 173 )

	224	Deep Learning Methods for Image Analysis and Synthesis for Intensity Modulated Radiotherapy:a Review
		Liu Guocai, Gu Dongdong, Liu Xiao, Liu Jinguang, Liu Yanfei, Zhang Maodan
		DOI: 10.3969/j.issn.0258-8021.2022.02.011
		Cancer is a common problem that seriously threatens human health. 60% to 70% of cancer patients need radiotherapy. Currently intensity modulated radiotherapy (IMRT) is one main radiotherapy technique that is widely applied in clinics. This paper reviewed deep learning methods, key technologies and future directions for IMRT, including clinical CT/CBCT/MRI/PET-guided IMRT technologies, and supervised or unsupervised deep convolutional neural networks or generative adversarial networks for the segmentation, registration and image-to-image translation of CT/CBCT/MRI/PET images of tumors.
		2022 Vol. 41 (2): 224-237 [Abstract] ( 194 ) HTML (1 KB) PDF (878 KB) ( 544 )

	238	Progress of Stem Cell Therapy in Interventional Regenerative Medicine
		Guan Siwen, Liu Xuan, Liu Gang
		DOI: 10.3969/j.issn.0258-8021.2022.02.012
		Interventional regenerative medicine (IRM) is a cross-discipline that achieves repair and regeneration of damaged organs through image-guided minimally invasive surgery and locoregional delivery of stem cells. In recent years, stem cell-based regenerative therapy has shown great promise in the preclinical setting, however, most intravenously delivered cells become trapped in the lungs and reticuloendothelial system, resulting in little cell reaching target tissues, so that the clinical results have been still suboptimal. IRM aims to increase the efficacy of stem cell-based regenerative therapies by locoregional targeted delivery of stem cell via endovascular, endoluminal, or direct injection into tissues and organs. In this review, we summarized the stem cell sources in regenerative therapies, the disease classifications, routes of delivery and mechanisms of action involved in the targeted delivery of stem cells. We showed the development of stem cell therapy in IRM. The application prospects and the barriers to translation of IRM were also discussed.
		2022 Vol. 41 (2): 238-246 [Abstract] ( 145 ) HTML (1 KB) PDF (2046 KB) ( 373 )

Communications

	247	Quantitative Analysis of Primary Cilia Based on Digital Image Processing
		Li Jin, Yang Pengfei, Wang Jufang, He Jinpeng, Ma Wei, Zhou Heng
		DOI: 10.3969/j.issn.0258-8021.2022.02.013

		2022 Vol. 41 (2): 247-251 [Abstract] ( 238 ) HTML (1 KB) PDF (4164 KB) ( 268 )

	252	The Influence of Cavity and Bony Structure on Dosimetry Evaluation in Nasopharyngeal CarcinomaIMRT Plans Using Different Algorithms
		Shen Zhengwen, Tan Xia, Li Shi, Tian Xiumei, Luo Huanli, Jin Fu, Wang Ying
		DOI: 10.3969/j.issn.0258-8021.2022.02.014

		2022 Vol. 41 (2): 252-256 [Abstract] ( 154 ) HTML (1 KB) PDF (741 KB) ( 389 )

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