Chinese Journal of Biomedical Engineering

Home

Editorial Board

Download

Quick Search

Office Online

Current Issue

Download Articles

Download

Links

2016 Vol. 35, No. 6 Published: 2016-12-20

Papers Reviews Communications

Papers

	641	A Novel Computer Aided Alzheimer′s Analysis Approach Based on Regions of Interests of PiB PET Images
		Duan Huoqiang Shu Xinghui Xu Jun Jiang Jiehui
		DOI: 10.3969/j.issn.0258-8021.2016.06.001
		Alzheimer′s disease (AD) is an irreversible neurodegenerative disease. PiB PET imaging can be used for AD diagnosis in the early stage. However, AD clinical diagnosis based on PiB PET technology is mainly based on visual assessment currently, which relies on the subjective experience of radiologists, leading to inaccurate and inadequate results. This study therefore proposed a computer aided AD diagnosis (CAAD) approach. The CAAD approach applied 3D thresholding-lattice boltzmann model foramyloid-beta (Aβ) ROIs segmentation, PCA method for feature extraction, and SVM with polynomial kernel for AD classification. In order to verify the CAAD approach, comparison experiments were carried out by selecting a total of 149 PiB PET datasheets from ADNI database and PET center of HuashanHospital (Shanghai). Results showed that the average accuracy rate of ROIs segmentation was 91.53%±3.0% in term of Dice coefficient. Besides, the classification accuracy of the novel CAAD approach for AD-healthy control (HC), mild cognitive impairment (MCI)\\|HC and AD/MCI-HC achieved 87.01%, 93.04% and 91.95%, respectively. Compared with the AD diagnosis methods in literatures, the CAAD approach could achieve a better accuracy of about 10% higher. These results indicated that the CAAD approach proposed in this study achieved the excellent performance for AD, MCI and HC classification.
		2016 Vol. 35 (6): 641-647 [Abstract] ( 835 ) HTML (1 KB) PDF (2617 KB) ( 686 )

	648	A Modified FCM Clustering Method for Brain Magnetic Resonance Image Segmentation
		Lin Xiangbo Wang Xinning Guo Dongmei
		DOI: 10.3969/j.issn.0258-8021.2016.06.002
		The noise and bias field are main factors lowering the quality of the magnetic resonance imaging. In order to segment brain tissue from MRI image, an anti-noise coherent local intensity fuzzy clustering algorithm (ANCLIFC) was proposed in this wok. By adding a new fuzzy operator and coherent local information as constraints in the cost function, ANCLIFC algorithm exhibited good clustering performance in resisting noise and bias field simultaneously. Twenty synthetic images, 20 simulated brain MRI images from BrainWeb and 100 real brain MRI images from IBSR database were used to evaluate the algorithm′s clustering performance. The experimental results demonstrated that ANCLIFC algorithm had better classification accuracy and stability than other classical modified FCM algorithms for low quality images contaminated by noise and bias field. For synthetic images, the average overall classification accuracy′s SA was 0.97, larger than other algorithms and the best improvement achieved 0.37. For real brain MRI images, ANCLIFC algorithm exhibits obvious superiority in segmenting CSF and the similarity measure′s KI increases about 0.1 in average.
		2016 Vol. 35 (6): 648-657 [Abstract] ( 547 ) HTML (1 KB) PDF (11007 KB) ( 287 )

	658	A Tongue Image Recognition Method Based on Type II Diabetes Traditional Chinese Medicine Syndrome Classification
		Kan Hongxing Zhang Luyao Dong Changwu
		DOI: 10.3969/j.issn.0258-8021.2016.06.003
		Tongue diagnosis is one of the methods oftype II diabetes traditional Chinese medicine (TCM) syndrome differentiation. In order to reduce the diagnostic deviation caused by external light and the doctor′s experience, a tongue image recognition method was proposed in this paper according to the TCM for syndrome classification of type II diabetes. Firstly, the corrected tongue image was obtained by color-calibration. Secondly, based on the difference between the color space, tongue image was divided into background and tongue area using fast K-mean clustering method,and then tongue area was separated into tongue body area and tongue fur area using Ohta color threshing method. Thirdly, according to the type IIdiabetes TCM tongue features, the color features, crack features and dental indentations features were extracted form tongue body area, tongue fur and tongue nature area. Finally, the random forest was used as the classifier to train tongue features data, and compared with the support vector machine method. The proposedmethod has been tested in 218 tongue image cases. Experimental results showed that the average recognition accuracy of random forest methods was 90.37%, which was increased by 10.74% compared with that of the support vector machine method, suggesting the random forest method was more accurate and efficient for type II diabetes TCM syndrome classifications.
		2016 Vol. 35 (6): 658-664 [Abstract] ( 580 ) HTML (1 KB) PDF (2681 KB) ( 661 )

	665	Magnetoencephalography Analysis of Schizophrenia Using Multiscale Permutation Entropy
		Chen Zhenyu Huang Xiaoxia
		DOI: 10.3969/j.issn.0258-8021.2016.06.004
		Multiscale permutation entropy was utilized to analyze complexity measure of resting state MEG recordings from schizophrenic patients and healthy control subjects in this paper. By comparing and analyzing the mean value of multiscale permutation entropy between schizophrenia and control subjects, we found that multiscale permutation entropy of the schizophrenia′s MEG in 93.82% channels was higher than that of the control subjects. In addition, the experimental result showed that multiscale permutation entropy of schizophrenics′ MEG recordings had significant differences in 19 channels (P<0.05, FDR correction). These channels were mainly distributed on the temporal and frontal, especially the edge region of temporal, which was consistent with the results of previous schizophrenia neuroimaging studies such as MRI and EEG. These features can be served to provide important reference for the further researches of schizophrenia pathological mechanism analysis and clinical diagnosis method.
		2016 Vol. 35 (6): 665-670 [Abstract] ( 443 ) HTML (1 KB) PDF (3204 KB) ( 511 )

	671	Synchronization in Alcoholics Using Cortical EEG
		Liu Guiqing Cao Rui Xiang Jie
		DOI: 10.3969/j.issn.0258-8021.2016.06.005
		Synchronization can measure the interaction between different brain regions. However, existing synchronization analyses on EEG synchronization are almost all based on the scalp EEG data. Due to the volume conductor effect, pseudo relevance between the scalp electrodes may occur, which further affects the measurement of synchronization. In this study, we selected the 61 channel EEG time series data of both the control subjects (28) and the alcoholics (28). For each trial, the subject was exposed to one of the following three different stimuli: a single stimulus (S1), two matched stimuli (S2 match), or two non-matched stimuli (S2 no match). The study used eLoreta and the self-defined 61 ROI (regions of interested) in the cerebral cortex. The synchronization between ROIs and the synchronization difference between alcoholic subjects and controls (non-alcoholics) during certain cognitive tasks were measured. Results indicated that the synchronization for the alcoholic group was lower than in the control group when performing the same cognitive task, the difference was statistically significant with α, β1 and θ band (P<0.05). The synchronization for the control group can reflect the complexity levels of the cognitive tasks (S2 match > S2 no match > S1), whereas the alcoholics only displayed erratic changes when the different cognitive tasks performed. The brain regional MSL between different subjects indicated that the alcoholic group was lower than control group when the S2 match performed,and in α band, frontal, left temporal area, central and right temporal area have significant differences (P<0.05); in β1 band, frontal showed significant differences (P<0.05); in β2 band, both frontal and left temporal area showed significant differences (P<0.05). These results indicate that in alcoholics, brain synchronization was reduced, and the connection between different brain regions was inhibited, meaning brain cognitive function is impaired. Our research provides indicated that long-term alcoholism causes functional damage to the brain from an entirely new perspective.
		2016 Vol. 35 (6): 671-676 [Abstract] ( 668 ) HTML (1 KB) PDF (0 KB) ( 178 )

	677	Ultrasound Beamforming Based on GPU Parallel Computation
		Chen Yinran He Qiong Luo Jianwen
		DOI: 10.3969/j.issn.0258-8021.2016.06.006
		The conventional line-by-line focused mode restricts the improvement of frame rate in ultrasound imaging. In plane wave imaging, each image is obtained by only one transmission and reception, and thus ultrafast imaging can be achieved. However, the current beamformers cannot meet the demand of ultrafast ultrasound imaging for massive computation. In this paper, the feasibility analysis of parallel computation in delay\\|and-sum beamforming algorithm was performed and two beamforming methods based on graphics processing unit (GPU) parallel computation including the 2 Kernel-based and 1 Kernel-based parallel beamformers were designed and implemented. The main differences between these methods were the calculations and data storage strategies of time delays in beamforming. Phantom experiments demonstrate that the computational frame rates of each method was 2,178 frames per second and 2,453 frames per second, respectively. Each of the methods obtained a speed up ratio of 99 and 111 compared to the normal method, which demonstrated that the GPU-based beamformer could significantly improve the calculating capability.
		2016 Vol. 35 (6): 677-683 [Abstract] ( 916 ) HTML (1 KB) PDF (10452 KB) ( 245 )

	684	The Construction and Analysis of Brain Functional Network on theEEG Response of Magnetic Stimulation at Guangming Point
		Xing Yangyang Geng Yuehua Zhang Xin Yu Hongli Xu Guizhi
		DOI: 10.3969/j.issn.0258-8021.2016.06.007
		Recently, the research and application on the combination of transcranial magnetic stimulation(TMS) and acupoint stimulation by traditional Chinese medicine has made some progress, graph theory has also been explored in the field of brain research. The combination of these theories and techniques will bring outa new direction for the application of neural magnetic stimulation and research of brain science. In this paper, the linear time domain correlation analysis of EEG between two channels of the magnetic stimulation at Guangming point and the mock point were carried out, each channel of the diencephalon electric data related to the number and the form of correlation matrix was determined by setting the size threshold for obtaining adjacency matrix. The brain functional network was constructed by the signals of the two experimental states. Results showed that the brain function of the laser beam was significantly increased than that of the occipital lobe (visual contact area) and the frontal area. Based on the graph theory, complex network theory was used to compare the parameters of the characteristic parameters of the brain function network.Magnetic acupuncture point laser made the brain functional network changed, the average degree of network and clustering coefficient were increased, and the characteristic path length decreased; meanwhile, the global efficiency and local efficiencywere improved, the property of small world was enhanced, all of which were statistically different.This work provided a theoretical basis for the application of magnetic stimulation technology in improving the nerve function of acupoints.
		2016 Vol. 35 (6): 684-690 [Abstract] ( 577 ) HTML (1 KB) PDF (1564 KB) ( 704 )

	691	Study on Multi-Mode Calculation Model in Non-Invasive Blood Pressure Measurement by Pulse Wave Velocity Method
		Gu Yaxiong Yang Tao Bao Ke Zhong Xinxin
		DOI: 10.3969/j.issn.0258-8021.2016.06.008
		Aiming to solve the problem of modeling difficulties and low calculating accuracy in the process of blood pressure measurement by pulse wave velocity method, a new multi modulus model of blood pressure calculation was built up with combination of multiple modulus T_PTT, ln(T_PTT) and (1/T_PTT)². Firstly, the parameters for the model were determined based on experimental data drawn from measurement carried out to 99 random voluntary subjects while performance evaluation indexes of each model were calculated simultaneously. It was found that the fitting correlation coefficient of the modulus model, equal to 0.891, was larger than that of any other model with the least error variance of the modulus model, equal to 6.1, smaller than that of any other model, which showed that the multi modulus model possessed better fitting effect and lower calculation error compared with single modulus models. Then, another 36 subjects' SBP and DBP data were collected by using mercury blood meter and the self-designed multi modulus blood pressure system separately, and the relevant parameters between the two methods were calculated. For SBP, d<6 mmHg, E_d=0.55 mm Hg and δ_d=2.98 mmHg, while for DBP, d<6 mmHg, E_d=0.57 mmHg and δ_d=3.42 mmHg, which could meet the requirements included in AAMI SP10\\|199. At last, the Bland-Altman difference method was applied to test the consistency of data between the two methods. A conclusion was drawn that 95% consistency limits of DBP and SBP were (-5.3, 6.4) and(7.2, -6.2) respectively, which satisfied the clinical demand and provided an evidence to the effectiveness of application of multi modulus model of blood pressure calculation in noninvasive blood pressure measurement. The research result proves that it is possible for multi modulus model of blood pressure calculation to be applied to noninvasive blood pressure measurement by pulse wave velocity method.
		2016 Vol. 35 (6): 691-698 [Abstract] ( 521 ) HTML (1 KB) PDF (1563 KB) ( 640 )

	699	Soft Tissue Modeling Based on MLS-MWS Mixing Method
		Chen Weidong Liu Bo Zhu Qiguang
		DOI: 10.3969/j.issn.0258-8021.2016.06.009
		Targeting to problem of the large computation amount of mesh-free model and tedious boundary processing, MLS-MWS based on the improved least square approximation was proposed in this work. At first, the model was divided into the internal domain and external domain based on the field theory. In the internal domain, nodes far from the natural boundary constructed the equation of equilibrium with the mesh-free method. The method has no numerical integration, which may reduce the computation amount. In the external domain, the boundary problem was weakened through the mesh-free partial weak method. Secondly in the strong method, second-order derivative of the displacement should be indicated with the least square approximation equation. Targeting tothe large computation amount of mesh-free model and low efficiency, it was proposed to improve the moving least square approximation function and derivate, and MLS derivative interpolation was carried out for the node value, for working out the second-order derivative and further simplifying the computation. Eventually, deformation simulation experiment of stretching and squeezing the liver model was carried out with the PHANTOM touch interactive equipment. According to the results, the single-step execution time of MLS-MWS combined model was reduced to 13.8 ms from the original 17.3 ms of the mesh-free model, while the frequency was improved from 56.80 frames/s to 72.46 frames/s. The mixed model integrated advantages of the mesh-free weak-strong method and dual mesh-free point-allotting method, which improved the instantaneity of the model effectively.
		2016 Vol. 35 (6): 699-704 [Abstract] ( 374 ) HTML (1 KB) PDF (1479 KB) ( 341 )

	705	Research of the CH-VAD Left Ventricular Assist Device in vivo
		Zhang Dong Yang Baiqing Chen Haibo Wang Hongyue Yuan Weimin Yang Jianzhong Jia Liujun Luo Fuliang Yang Peng Tang Yue
		DOI: 10.3969/j.issn.0258-8021.2016.06.010
		To evaluate the security and reliability of a magnetic left ventricular assist device(CH-VAD),an in vivo experimental investigation was performed in the ovine model. A total of six healthy male sheep were involved in the experiment. Under anesthesia via lateral thoracotomy and ventilator assisted breathing, each of the 6 healthy male sheep was performed the CH-VAD implantation between the left ventricular apex and descending aorta under beating-heart condition. Postoperative nursing and anticoagulating therapy were completed routinely, and the operating parameters of CH-VAD were recorded continuously. The hematologic and biochemical tests were performed preoperatively and postoperatively to evaluate primary organ functions. Operating parameters of CH-VAD were recorded continuously until termination of the experiment.At the end of the experimental period, the sheep was humanely killed,and the end-organs were examined macroscopically and histopathologically.Two of the 6 animals died of ventricular fibrillation or asphyxia. One animal died at the 20th day because of respiratory failure, and 2 sheep survived to the end of the experiment (60 days). The other one animal died at the 135th day because of respiratory failure. For the surviving animals, hematological and biochemical data showed no evidence of organ dysfunctions during the experimental period, and the CH-VADs were in a good condition. At autopsy,histopathologic examination revealed that two sheep surviving for 20 and 135 days showed lobular pneumonia and multiple infarction of kidney, and another two sheep survivingfor 60 days showed local infarction of kidney. However the other end-organs did not have obvious pathological changes.Our study provided a theoretical and technical support for the preclinical evaluation and clinical operation of CH-VAD, and proved that CH-VAD had a relatively stable performance; proposed the thought and idea for building the standard of preclinical evaluation of left ventricular assist device, and provided the research direction on the subsequent implantation and clinical operation.
		2016 Vol. 35 (6): 705-711 [Abstract] ( 756 ) HTML (1 KB) PDF (5169 KB) ( 674 )

	712	IMRT Plan Optimization Based on Hybrid Criteria
		Guo Caiping Shu Huazhong Gui Zhiguo Zhang Liyuan
		DOI: 10.3969/j.issn.0258-8021.2016.06.011
		In optimization methods of conformal intensity modulated radiation therapy, the performance of biological optimization based on generalized equivalent uniform dose (gEUD) still requires improvement to control the target dose coverage precisely, while physical optimization based on dose volume does not reflect the nonlinear response of tissue to dose. Hence, a hybrid criteria optimization method integrating the biological criteria (generalized equivalent uniform dose: gEUD) and physical criteria (minimum dose, mean dose) was proposed in this paper. The new algorithm,taking full advantages of these two kinds of criteria, gave consideration to both the dose coverage of the target area and the protection of the organ.Its feasibility was tested on ten prostate cases through evaluation and comparison from the perspective of dosimetry and biology. Compared with physical criteria optimization, the hybrid criteria optimization reduced dose to the organs at risk on the premise that dose coverage characteristics of target were similar, and at the significance level of 0.05, the mean dose for rectum, V₅₀ and V₆₀ of rectum, the mean dose for bladder, V₆₅,V₇₀,V₇₅,NTCP and gEUD of bladder were significantly different (P<0.05). Moreover, compared with gEUD based biological optimization, on the one hand the target dose coverage characteristics have been greatly improved with dose statistics, and the biological indicators were significantly different (P<0.05); on the other hand, organs at risk got better protection with significant difference (P<0.05) in rectal average dose, V₅₀, V₆₀, V₇₅, NTCP and gEUD as well as in bladder V₇₅ and gEUD. In conclusion, gEUD-based hybrid criteria optimization could reduce the dose to OAR that may be helpful to further improve the dose coverage of PTV and to increase the gain ratio of radio therapy while guarantying the dose to PTV.
		2016 Vol. 35 (6): 712-718 [Abstract] ( 435 ) HTML (1 KB) PDF (970 KB) ( 457 )

	719	An Inducible Bone Repair Composite Material PHBV/CS: The Renovation of Rabbit Radius
		Li Gen Zhen Ping Cheng Manping Zhao Hongbin
		DOI: 10.3969/j.issn.0258-8021.2016.06.012
		In this work, four kinds of chitosan/poly hydroxybutyrate-co-hydroxyvalerate(PHBV)composite scaffold were fabricated by the method of two phase mixture freeze drying.The repair effects of the scaffoldswere investigated and compared to determine the optimal formula for guiding bone regeneration as well as drug releasing. The icariin (Ica) was used in this study as an osteoinductive factor, and Ica microspheres were fabricated by W/O method and covalent binding. Microstructures of the scaffolds and drug releasing behaviors were investigated.After that,the scaffolds were implanted in the radius defects of rabbits;X-ray and 3D CT were applied to observe the repair effect at 1, 3, and 6 months post surgery. The osteogenesis inducible effects were evaluated using HE and Masson staining. The scaffolds showed network-like microstructures with particles inside, the drug loaded particles was 3-11 μm in diameter. The scaffolds released Ica well, though the peak of drug release curve for the covalent binding group was delayed than the others. The peak appeared at 72 h after immersion and quickly reached a stabled level of 75 μg. The images of X-ray and 3-D CT showed the defect of the rabbit radius had connected and the bone mineral density was higher than the other three groups. HE and Masson stained sections of the radius segmental defect of rabbits implanted for 1, 3, and 6 months showed that the scaffold with covalently binding Ica had better repair effect than the others.
		2016 Vol. 35 (6): 719-728 [Abstract] ( 344 ) HTML (1 KB) PDF (33009 KB) ( 140 )

Reviews

	729	Progress of Biological Magnetoacoustic Tomography with Magnetic Induction
		Ma Zhen Sun Zheng Wang Jianjian
		DOI: 10.3969/j.issn.0258-8021.2016.06.013
		Biological magnetoacoustic tomography with magnetic induction (MAT-MI), a newly emerged functional imaging modality, combines the electromagnetic, ultrasonic and multi-physics detection and imaging technology. It has the advantages of good contrast of electrical impedance tomography and high spatial resolution of ultrasonography. It provides the information of conductivity changes of biological tissues and is helpful in the early diagnosis of pathological biological tissues. Endoscopic MAT-MI (EMAT-MI) combines MAT-MI and endoscopic detection techniques. It can directly detect the physiology and pathology of the luminal structures including nasal cavity, digestive tract and vessel. In this paper, the feasibility and difficulties of EMAT-MI were discussed based on the review of MAT-MI forward/inverse problem.
		2016 Vol. 35 (6): 729-736 [Abstract] ( 619 ) HTML (1 KB) PDF (1270 KB) ( 690 )

	737	Research Progress on Child Head Finite Element Model
		Cui Shihai Chen Yue Li Haiyan Ruan Shijie
		DOI: 10.3969/j.issn.0258-8021.2016.06.014
		Head injury is the main factor that leads to children′s death or disability. Therefore, research on children′s head injury biomechanics is of great significance. Recently, more and more investigations focus on using finite element child models to study the head injury, which can replace cadaveric experiments, animal experiments and physical experiments to some extents. This paper systematically summarized the progress of child head finite element model from the views of child age characteristics, model developingmethods and model application. Also, further research aspects and topics on finite element models were also proposed.
		2016 Vol. 35 (6): 737-743 [Abstract] ( 483 ) HTML (1 KB) PDF (5905 KB) ( 549 )

Communications

	744	An Improved Algorithm for Removing the Power-Line Interference from ECG Signals in High Sampling Rate
		Wang Jiming Lv Yingying Dong Han Wu Tao Bao Tao Feng Jingyi
		DOI: 10.3969/j.issn.0258-8021.2016.06.015

		2016 Vol. 35 (6): 744-748 [Abstract] ( 444 ) HTML (1 KB) PDF (978 KB) ( 688 )

	749	Brain Functional Networks under Different-Frequency Magnetic Stimulation at Neiguan Acupoint Based on EEG
		Yu Hongli Xu Guizhi Fu Lingdi Zhai Yue
		DOI: 10.3969/j.issn.0258-8021.2016.06.016

		2016 Vol. 35 (6): 749-753 [Abstract] ( 459 ) HTML (1 KB) PDF (3515 KB) ( 522 )

	754	Development of Panel Breast PET Based on Coupling of BGO and SiPM
		Chen Yanhua Xie Liang Yang Yong Yuan Xiabing Hou Liqun Li Yi Zhao Yongjie Wang Yuanji
		DOI: 10.3969/j.issn.0258-8021.2016.06.017

		2016 Vol. 35 (6): 754-758 [Abstract] ( 446 ) HTML (1 KB) PDF (5340 KB) ( 500 )

	759	Design of the Bladder Pressure Monitoring System Based on Wireless Passive Technology
		Li Chunchan Zhang Xu Li Gang Liu Bin Qian Bolin
		DOI: 10.3969/j.issn.0258-8021.2016.06.018

		2016 Vol. 35 (6): 759-763 [Abstract] ( 432 ) HTML (1 KB) PDF (4222 KB) ( 583 )

	764	Visualization Method and Knowledge Discovery of Prescription Composition
		Fan Fengjie Hong Wenxue Song Jialin Li Shaoxiong Zheng Cunfang
		DOI: 10.3969/j.issn.0258-8021.2016.06.019

		2016 Vol. 35 (6): 764-768 [Abstract] ( 320 ) HTML (1 KB) PDF (2955 KB) ( 464 )

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