Chinese Journal of Biomedical Engineering

Home

Editorial Board

Download

Quick Search

Office Online

Current Issue

Download Articles

Download

Links

2019 Vol. 38, No. 1 Published: 2019-02-20

Reviews Communications Regular Papers

Regular Papers

	1	Blood Vessel Segmentation of Fundus Images Based on Improved U Network
		Gao Hongjie, Qiu Tianshuang, Chou Yuanting, Zhou Ming, Zhang Xiaobo
		DOI: 10.3969/j.issn.0258-8021.2019.01.001
		The blood vessel segmentation of fundus images is the basis of computer-aided diagnosis of ophthalmology and other related diseases. Early diagnosis and monitoring of diseases such as diabetic retinopathy, hypertension and arteriosclerosis can be performed by segmenting the vascular structure in the fundus image. However, existing segmentation algorithms are challenged with low accuracy and low sensitivity. This paper proposed an improved U-Net fundus image segmentation algorithm based on the basic theory of deep learning. Firstly, the problem of less fundus data set was solved by reducing the number of pooling layers and upsampling layers of the traditional U-Net. Secondly, the use efficiency of the feature was improved by changing the traditional convolutional layer serial connection method to the residual mapping. Finally, the batch normalization and PReLU activation functions are added between the convolutional layers to optimize the network, which further improved the network performance. This paper conducted experiments on two public fundus databases, DRIVE and CHASE_DB1. The 160 000 image blocks were randomly extracted from each database and are sent into the improved network for training and testing. The sensitivity, accuracy and AUC (area under the ROC curve) of the algorithm were 2.47%, 0.21% and 0.35%, higher than those of the existing contrasted algorithms. The proposed algorithm improves the low accuracy and low sensitivity of small blood vessels segmentation in fundus images, and segmented small blood vessels better with low contrast.
		2019 Vol. 38 (1): 1-8 [Abstract] ( 746 ) HTML (1 KB) PDF (5662 KB) ( 769 )

	9	Optic Disc Localization Based on Regional Proposal Strategy
		Tang Yiping, Wang Liran, He Xia, Chen Peng, Yuan Gongping
		DOI: 10.3969/j.issn.0258-8021.2019.01.002
		The localization of optic disc (OD) is very important for computer-aided diagnosis of the ophthalmology diseases with fundus images. In this paper, a method of OD localization based on regional proposal strategy was proposed. First, the fundus image was mapped from the pixel domain to the feature domain, and candidate regions of OD were generated by using the regional proposal strategy in the obtained feature maps. Next, the candidate regions were sampled according to the certain criteria, and a fully connected layer was constructed to perform deep feature extraction. The location refinement of the candidate region was achieved by using the constraint of the loss function. At last, OD visibility was judged by filtering of the confidence threshold. If the OD was visible, the center of the candidate region with the highest degree of confidence was regarded as OD coordinate of the fundus image. The correct position of OD was obtained. Experiments were conducted in three public fundus image databases (DRIVE (40 images), MESSIDOR (1200 images) and STARE (400 images)). Testing results were 100%, 99.9% and 98.8%. Experimental results showed that the proposed method could reach the OD localization fast, accurately and robust, which was superior to existing OD localization methods. The pre-judgment of OD visibility was more consistent with the requirements of practical application. The proposed method was expected to contribute to the diagnosis of fundus diseases.
		2019 Vol. 38 (1): 9-17 [Abstract] ( 560 ) HTML (1 KB) PDF (8525 KB) ( 255 )

	18	A Three-Dimensional Liver Vessel Segmentation Method for CT Images Using Improved Fuzzy Connectedness
		Zhang Rui, Wu Weiwei, Zhou Zhuhuang, Jiang Tao, Wu Shuicai
		DOI: 10.3969/j.issn.0258-8021.2019.01.003
		Traditional fuzzy connectedness methods exist some drawbacks in segmentation of liver vessels from computed tomography (CT) images, including unsatisfactory segmentation performance, requirement on multiple seeds, and low time efficiency. In this paper, the traditional fuzzy connectedness method was improved from following three steps: 1) The Jerman′s vesselness filter was improved; 2) The improved vesselness was incorporated into the fuzzy affinity function; 3) The fuzzy connectedness was initialized by the Otsu multi-thresholding algorithm instead of the confidence connectedness. The preprocessing comprised adaptive sigmoid filtering and isotropic resample filtering. Next, the improved Jerman′s vesselness filter was performed. Then, the improved Jerman′s vesselness was integrated into the fuzzy affinity function. The foreground information was analyzed to initialize the fuzzy connectedness by using the Otsu multi-thresholding algorithm. Finally, three-dimensional (3D) liver vessels were segmented with one single seed. The improved vesselness filter and the improved fuzzy connectedness method were quantitatively evaluated by using 20 cases of public CT data sets. The evaluation metrics included contrast to noise ratio (CNR), accuracy, sensitivity and specificity. The average CNR of the improved vesselness filter was 8.43 dB,which was superior to the traditional vesselness filters. The accuracy of the proposed vessel segmentation method was 98.11%, which was better than the traditional fuzzy connectedness method based on confidence connectedness and the regional growing and level set methods. In addition, the proposed method also had advantages in terms of time efficiency. The 3D segmentation method proposed in this paper could effectively address the issues associated with the traditional fuzzy connectedness method and improve the accuracy and efficiency of 3D liver vessel segmentation in CT images.
		2019 Vol. 38 (1): 18-27 [Abstract] ( 503 ) HTML (1 KB) PDF (10886 KB) ( 180 )

	28	A Novel Automated Tumor Segmentation Model for Enhanced Breast MRI
		Ma Wei, Liu Hongli, Sun Mingjian, Xu Jun, Jiang Yanni
		DOI: 10.3969/j.issn.0258-8021.2019.01.004
		Breast cancer can be mainly classified into two kinds: mass-like and non-mass-like on enhanced breast images. Owing to the small area of breast cancer, along with the huge difference between the shape of mass-like and non-mass-like and the self complexity of non-mass-like, it is hard to segment the accurate area of breast tumor. To solve these problems, this paper proposed a novel deep learning model of rough detection and fine segmentation. Before precise segmentation, rough detection for the cancer region was firstly processed for potential region of the tumor. On the basis of rough detection, we used SegNet for fine segmentation to achieve the best performance of the algorithm. In order to test the effectiveness of proposed method (YOLOv2+SegNet), we picked 560 magnetic resonance imaging (MRI) images of breast cance out of the dataset collected from the hospital for training and testing (415 images for training and 145 for testing). For more comprehensive analysis, experiments were set to analyze three different conditions, such as mass-like, non-mass-like and the mix of mass-like and non-mass-like. From the results, the established method improved 10% under each condition and improved a lot compared with the traditional C-V model, fuzzy C mean clustering, active contour model for spectral mapping and deep model of U-net or DeepLab.
		2019 Vol. 38 (1): 28-34 [Abstract] ( 549 ) HTML (1 KB) PDF (4404 KB) ( 773 )

	35	Chest Electrical Impedance Tomography Method Based on Priori Information of Human Body Structure
		Wang Qi, Chen Xiaojing, Wang Jianming, Li Xiuyan, Duan Xiaojie, Wang Huaxiang
		DOI: 10.3969/j.issn.0258-8021.2019.01.005
		Electrical impedance tomography (EIT) technique has important clinical values in human thoracic pathological changes and lung detection. Due to the specificity of the chest contour, the reconstructed images based on traditional model imaging methods often have large errors. In this paper, we proposed a chest electrical impedance tomography method based on prior information of human body structure. The contours of the chest and lungs were extracted through the image processing of CT images, which provided prior information for forward and inverse problems of EIT. At the same time, an efficient subdivision method for inverse problem was proposed, which makes the shapes of reconstructed images closer to the real one. As a result, the quality of reconstruction was improved. In order to verify the effectiveness of the method, thirty samples of lung CT images for healthy human were selected from a hospital CT database. For the proposed method and two traditional methods, namely elliptical model imaging method and circular model imaging method, the statistical analysis of the lung region ratio (LRR) for the three methods were conducted. The results showed that there was no significant difference between the real LRR and the computed LRR based on the proposed method. The relative errors between the computed LRR based on proposed method and the real one was 3.71%±1.77%, which was much smaller than the elliptical model imaging method (10.29%±3.30%) and the circular model imaging method (12.74%±2.87%). The statistical significance was P<0.05. In conclusion, the proposed method could effectively improve the imaging quality.
		2019 Vol. 38 (1): 35-43 [Abstract] ( 452 ) HTML (1 KB) PDF (8550 KB) ( 161 )

	44	Effect of Stimulation Type on Auditory Steady-State Response
		Wang Jinhai, Jia Yaru, Chen Xiaogang, Wang Yao, Li Kun, Meng Jia
		DOI: 10.3969/j.issn.0258-8021.2019.01.006
		Auditory steady state response (ASSR) is an electroencephalography (EEG) potential elicited by the periodic auditory stimuli, which can be used for building auditory brain-computer interface (BCI). In order to build efficient ASSR-based BCIs, it is necessary to seek a stimulation type to evoke stronger ASSR. In this work the canonical correlation analysis (CCA) was used to compare the ASSRs from 14 healthy subjects evoked by click, sinusoidal amplitude modulation (SAM), and white noise. Results indicated that the three stimulation types could evoke stable ASSR, and the strongest ASSR was mainly concentrated in frontal-central area. The strength of ASSR was depended on stimulation type. The strongest ASSR was obtained by the click stimuli. The second one and the weakest ASSR were corresponding to the white noise stimuli and SAM stimuli respectively. Using CCA to classify the left and right ears for the three stimulation types, we found out that the click stimuli induced the highest information transfer rate (ITR), which was 6.69 bits/min; the second one and the weakest ITR were corresponding to the white noise stimuli and SAM stimuli, which was 1.65 bits/min and 0.76 bits/min respectively. These results indicated that click stimuli was suitable for building high-speed BCI systems.
		2019 Vol. 38 (1): 44-50 [Abstract] ( 536 ) HTML (1 KB) PDF (2813 KB) ( 532 )

	51	A Study of Argument Reality Based Brain-Computer Interface (AR-BCI) in Hololens
		Zhang Lixin, Zhang Yukun, Ke Yufeng, Du Jiale, Xu Minpeng, Ming Dong
		DOI: 10.3969/j.issn.0258-8021.2019.01.007
		Brain-Computer interfaces (BCIs) have improved greatly in the last decades. However, high-performance BCIs usually need display devices to present the visual stimulus to evoke specific EEG patterns. The most popular display device now is computer monitor that is not portable and thus restrict the portability of BCI. By combining argument reality (AR) technology with BCIs, the problem can be resolved, achieving a more practically applicable BCI system. Currently, how to raise the speed and accuracy of AR-BCI remains an open question. This study proposed an AR-BCI system based on Hololens. We generated eight stimuli in the argument reality environment to evoke different steady state visual evoke potentials (SSVEP). Twelve subjects participated in this study and their SSVEP patterns were successfully evoked in AR environment. They achieved 88.67% and 98.6% accuracy on average for SSVEP-BCI with EEG data length of 1 s and 2 s respectively. Our Result showed that AR-BCI is promising to achieve a high-performance portable and wearable control system in daily life.
		2019 Vol. 38 (1): 51-58 [Abstract] ( 513 ) HTML (1 KB) PDF (4959 KB) ( 604 )

	59	Dysphonic Analysis of Parkinson′s Disease Based on Partially Ordered Topological Graph
		Zhang Tao, Jiang Peipei, Li Lin, Zhang Xiaojuan
		DOI: 10.3969/j.issn.0258-8021.2019.01.008
		In this paper, we proposed a novel dysphonic analysis method on Parkinson′s disease based on partially ordered topological graph from the view of formal concept analysis. Firstly, we constructed a representation (method) named partially ordered topological graph (POT graph) from attribute topology and attribute partially ordered graph, which gained the ability of concept searching and hierarchical concept tree structure representation. Coloring and briefing the concept tree could obtain the brief concept tree. The concept classification structure of the analysis object could be obtained according to the partial order relation of the brief concept tree. Applying the method to concept searching in Parkinson′s disease dataset, results showed that the POT graph could not only analyze the relationship between Parkinson′s disease and speech feature in the view of formal concept, but also be used as a diagnostic basis for data analysis. Results obtained from several Parkinson′s disease datasets(the numbers of sample are 197, 5875, 1040 and 220)showed that the average precision was 76.64% by POT graph. Compared with the classical classifier such as LDA (67.36%), QDA (70.83%), kNN (71.83%), Parzen window (70.24%), and SVM (74.61%), our result was higher than SVM 2.72%. In conclusion, the proposed method could be beneficial to the dysphonic analysis of Parkinson′s disease.
		2019 Vol. 38 (1): 59-69 [Abstract] ( 445 ) HTML (1 KB) PDF (3917 KB) ( 463 )

	70	Effect of Extracellular Matrix Protein Pattern on the Differentiation of Myoblast
		Qiu Changjun, Hun Tingting, Zhao Yingtong, Zhan Yuewei, Zhao Feng, Sun Yan
		DOI: 10.3969/j.issn.0258-8021.2019.01.009
		The differentiation mechanism of skeletal myoblast differentiation is still unclear. The main purpose of this research is to explore the effect of the extracellular matrix (ECM) protein pattern on myogenic differentiation. In this study, four different fibronectin micropatterns (parallel, random, vertical, and control group) were prepared by microcontact printing technology, and the differentiation of C2C12 cells after 7 day incubation was analyzed by immunofluorescence staining and analyzes the adhesion molecule, Vinculin and F-actin distribution after 2 h, 6 h, 18 h between groups by immunofluorescence staining method, each group experiment repeated 3 times, each time point had 3 parallel samples. The myotubes number formed in the control grouphad no significant difference compared with that inthe parallel alignment group(17.33±0.58 vs 16±1.73), but wassignificantly higher than the vertical group(7.00±1.00) and the random group significantly(10.89±0.19) (P<0.05). At 2 h, the arrangement of microfilaments in cells was consistent with the local micropatterns. At 6 h, the microfilament began to align along the long axis direction of the whole strips pattern. At 12 h, the distribution of microfilaments was consistent with that of the long axis direction of the whole strips pattern. At 18 h, the adhesion plaques formed was distributed along the micropattern area. In conclusion, the extracellular matrix protein patterns are influential on the differentiation of myoblasts, different fibronectin polarity micropatterns have distinct effects on myoblasts cytoskeleton and adhesion, and it might be one of the reasons that later differentiation of C2C12 results in different outcomes.
		2019 Vol. 38 (1): 70-76 [Abstract] ( 375 ) HTML (1 KB) PDF (6833 KB) ( 310 )

	77	Preparation and Hydrophilic Characterization of Monodispersed PEG-b-PCL/ PCL Microspheres by Electrospraying
		Yang Xue, Shao Huaying, Zhang Qiongyue, Wu Xiaohong
		DOI: 10.3969/j.issn.0258-8021.2019.01.010
		This study is aimed to prepare monodispersed and hydrophilic microspheres with Poly(ethylene glycol)-b-poly (ε-caprolactone) / Polycaprolactone (PEG-b-PCL/PCL) by electrospraying. PEG-b-PCL and PCL was dissolved in chloroform to prepare a mixture solution, which was magnetically stirred for 3 hours. The influence of the content of PEG-b-PCL, flow rateand the voltage of the electrospraying on the morphology, size, and size distribution of microspheres were investigated, The influence of PEG-b-PCL on hydrophilicity of microspheres and the dispersion of the microspheres in water was investigated. Monodispersed spherical microspheres with average particle sizes ranging from 5 to 6 μm and the coefficient of variation ranging from 15%～21% could be fabricated when the content of PEG-b-PCL was 10%～20%, flow rate was 1mL/h, and voltage was 10 kV. Microspheres along with fibers could be fabricated when the content of PEG-b-PCL was 30%. The contact angle of the microspheres decreased from 126.2°±4.8° to 29.9°±4.9° when the content of PEG-b-PCL was increased from 0 to 20%, and the differences showed statistical significance (P<0.05), indicating that changing the content of PEG-b-PCL could improve the hydrophilicity of the microspheres. In the meantime, microspheres with 10%～20% of PEG-b-PCL could form homogeneous dispersions in water. Conclusion: Monodispersed microspheres with improved hydrophilicity could be fabricated with PEG-b-PCL/PCL, which would be a foundation for further research on hydrophilic drug-loaded microspheres.
		2019 Vol. 38 (1): 77-83 [Abstract] ( 358 ) HTML (1 KB) PDF (4822 KB) ( 456 )

Reviews

	84	New Developments and Trends of BCI Based on Motor Imagery
		Zhao Xin, Chen Zhitang, Wang Kun, Wang Zhongpeng, Zhou Peng, Qi Hongzhi
		DOI: 10.3969/j.issn.0258-8021.2019.01.011
		Brain-computer interface (BCI) based on motor imagery (MI) is a new rehabilitation method, which plays a significant role in helping to improve and restore physical functions of patients. However, MI-BCI still faces many challenges for practical application, including the low spatial resolution of physiological signals induced by MI, the long training time of users, and the difficulty in implementing an asynchronous control MI-BCI system. This paper briefly outlined the research of MI-related mechanisms, reviewed the relevant solutions and the research status from the aspects of signal acquisition, signal processing algorithm analysis, paradigm design and asynchronous control research. At last, we outlined the application and perspectives of MI-BCI in the future development.
		2019 Vol. 38 (1): 84-93 [Abstract] ( 860 ) HTML (1 KB) PDF (851 KB) ( 1567 )

	94	A Review of Cognition Related Macro-Microstructural Changes Based on MRI
		Wu Qiong, Chen Yuanyaun, Zhao Xin, Ming Dong
		DOI: 10.3969/j.issn.0258-8021.2019.01.012
		Alzheimer′s disease (AD) is a neurodegenerative disease with cognitive impairment. The mechanism of AD is complex and not clear now and therefore still a great challenge in the field of medical research. The development of structure and diffusion magnetic resonance imaging technology can provide important technical means for studying the mechanism of macroscopic morphology and microstructural pathological characteristics of Alzheimer′s disease. In recent years, studies have observed that there are relationships between the abnormal changes of macroscopic and microscopic structure, which is of great significance for revealing the mechanism of structural changes in age-related diseases including AD. This article reviewed the studies on the pattern and interrelation of brain macro-microstructure changes to cognitive impairment in recent years, and summarized the important analysis methods and research conclusions. The mechanism of brain macro-microstructure pathology was discussed as well.
		2019 Vol. 38 (1): 94-101 [Abstract] ( 472 ) HTML (1 KB) PDF (804 KB) ( 594 )

	102	Pharmacogenomics Clinical Decision Support System:A Review
		Qin Weifeng, Lu Xudong, Duan Huilong, Li Haomin
		DOI: 10.3969/j.issn.0258-8021.2019.01.013
		In recent years, precision medicine has become an intensively researched issue in the field of biomedicine. Many countries in the world have been endeavoring to make breakthroughs in this field. Pharmacogenomics (PGx) which use of genomics and other “omics” knowledge to individualize drug selection and drug use to avoid adverse drug reactions and to maximize drug efficacy is an important part of precision medicine, as well as one of the most promising parts in the field of precision medicine to realize the clinical routine use of PGx and promote the progress of precision medicine. The pharmacogenomics clinical decision support (PGx-CDS) system is an essential tool for the clinical implementation and knowledge translation of pharmacogenomics. In many clinical institutions, pharmacogenomics services have been successfully implemented along with the deployment of the PGx-CDS system, and many clinical and research institutions are also preparing to conduct these services. In this paper, the main PGx-CDS systems that have been presented and reported in literatures are reviewed, 11 PGx-CDS systems are involved in total. Through a comprehensive review of the clinical application background, system design, knowledge representation, intervention mode and system evaluation aspects of these PGx-CDS systems, the research progress and development status of the current PGx-CDS systems are summarized, and the main challenges and the future development directions of the PGx-CDS system are discussed.
		2019 Vol. 38 (1): 102-111 [Abstract] ( 607 ) HTML (1 KB) PDF (1208 KB) ( 671 )

	112	Recent Advances in Preparation of Albumin Nanoparticle Systems
		Li Yan, Lan Jinxiao, Luo Cheng
		DOI: 10.3969/j.issn.0258-8021.2019.01.014
		In the past decades, drug delivery systems based on nanoparticles have been increasingly applied in medical fields, aiming to improve the target activity of drugs and reduce the untoward effects. Among various polymers, albumin is an ideal material for the formulation of nanoparticle-based drug delivery systems. As the most abundant protein in plasma, albumin has many advantages, such as good biocompatibility, biodegradability and lack of immunogenicity. There is a list of approaches to prepare albumin nanoparticles, including desolvation, self-assembly, emulsification, double emulsification, thermal gelation, spray drying, nab-technology and pH-coacervation. Due to the differences in the mechanism and conditions of preparation, each method has its own advantages and disadvantages. This article provided an overview of research progresses in different albumin-based formulations according to preparation methods. The current difficulties confronted by albumin nanoparticles and the future development directions were also discussed.
		2019 Vol. 38 (1): 112-119 [Abstract] ( 869 ) HTML (1 KB) PDF (2769 KB) ( 1803 )

Communications

	120	Research and Implementation of Wireless Brain-Computer Interface System Based on SSVEP
		Wu Zhengping, Wei Huan, Zhao Jing, YangXiangyu, Qiu Kai
		DOI: 10.3969/j.issn.0258-8021.2019.01.015

		2019 Vol. 38 (1): 120-124 [Abstract] ( 536 ) HTML (1 KB) PDF (3198 KB) ( 563 )

	125	Influence of Cross-Shear on the Wear of PEEK and CFR-PEEK in Artificial Joint Application
		Wang Junyuan, Duan Chenxi, Du Wenhua, Dong Lei
		DOI: 10.3969/j.issn.0258-8021.2019.01.016

		2019 Vol. 38 (1): 125-128 [Abstract] ( 393 ) HTML (1 KB) PDF (1696 KB) ( 338 )

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