Pub Date : 2009-10-30DOI: 10.1109/BMEI.2009.5305528
L. Wang, Jinwen Ma
The Gaussian mixture model is a powerful statistical tool in data modeling and analysis. Generally, the EM algorithm is utilized to learn the parameters of the Gaussian mixture. However, the EM algorithm is based on the maximum likelihood framework and cannot determine the number of Gaussians for a sample data set. In order to overcome this problem, we propose a new model selection criterion based on the kurtosis and skewness of the estimated Gaussians. Moreover, a new greedy EM algorithm is constructed via the kurtosis and skewness based criterion. The simulation results show that the proposed model selection criterion is efficient and the new greedy EM algorithm is feasible.
{"title":"A Kurtosis and Skewness Based Criterion for Model Selection on Gaussian Mixture","authors":"L. Wang, Jinwen Ma","doi":"10.1109/BMEI.2009.5305528","DOIUrl":"https://doi.org/10.1109/BMEI.2009.5305528","url":null,"abstract":"The Gaussian mixture model is a powerful statistical tool in data modeling and analysis. Generally, the EM algorithm is utilized to learn the parameters of the Gaussian mixture. However, the EM algorithm is based on the maximum likelihood framework and cannot determine the number of Gaussians for a sample data set. In order to overcome this problem, we propose a new model selection criterion based on the kurtosis and skewness of the estimated Gaussians. Moreover, a new greedy EM algorithm is constructed via the kurtosis and skewness based criterion. The simulation results show that the proposed model selection criterion is efficient and the new greedy EM algorithm is feasible.","PeriodicalId":6389,"journal":{"name":"2009 2nd International Conference on Biomedical Engineering and Informatics","volume":"49 1","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2009-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86630621","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-10-30DOI: 10.1109/BMEI.2009.5305838
Lei Wang, Jing Si, Xiaowei Hu
For the problem that the centralized mechanism for semantic service discovery is still difficult for realization in the server center during service discovery and matching, a novel model of web services is proposed which makes use of some ideas from the peer-to-peer structure and the clustering methods. In this model, a service is firstly matched within its located community in accordance with the semantic information for its function, and then a rather accurate matching is made according to the semantic information from input parameters. Both theoretical analysis and simulation results show that this model is able to raise the recall ratio as well as to improve the matching speed, and moreover, it can also provide users with value-added response from web services composition which may meet user’s further requirements. These features not only satisfy some user’s potential needs, but also deal with the problem of the centralized mechanism for semantic service discovery during the serviceoriented computation. Keywords-P2P; clustering; Web-service; services composition
{"title":"Research on the Clustering and Composition of P2P-Based Web Services","authors":"Lei Wang, Jing Si, Xiaowei Hu","doi":"10.1109/BMEI.2009.5305838","DOIUrl":"https://doi.org/10.1109/BMEI.2009.5305838","url":null,"abstract":"For the problem that the centralized mechanism for semantic service discovery is still difficult for realization in the server center during service discovery and matching, a novel model of web services is proposed which makes use of some ideas from the peer-to-peer structure and the clustering methods. In this model, a service is firstly matched within its located community in accordance with the semantic information for its function, and then a rather accurate matching is made according to the semantic information from input parameters. Both theoretical analysis and simulation results show that this model is able to raise the recall ratio as well as to improve the matching speed, and moreover, it can also provide users with value-added response from web services composition which may meet user’s further requirements. These features not only satisfy some user’s potential needs, but also deal with the problem of the centralized mechanism for semantic service discovery during the serviceoriented computation. Keywords-P2P; clustering; Web-service; services composition","PeriodicalId":6389,"journal":{"name":"2009 2nd International Conference on Biomedical Engineering and Informatics","volume":"71 1","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2009-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89188060","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-10-30DOI: 10.1109/BMEI.2009.5305065
Jianguo Zhang, Haiyan Song, Q. Xue, Taoran Liu, Fang Wang
By studying surface electromyography(sEMG) characteristics of human neck, shoulder and back muscles keeping static posture, to explore the precise evaluation index for muscle function of the human body, and the significance of muscle function prediction, diagnosis and rehabilitation evaluation. Method: 9 experiment objects, the sEMG signals of five muscles (neck extensor, upper trapezius, the middle trapezius, posterior deltoid, latissimus dorsi ) at deferent angles and deferent bed mattresses were tested by 8 channel sEMG collected system. The eigenvalue of MPF( Means power frequency)were processed by SPSS 16.0. Conclusion: deferent angles and deferent mattresses influenced greatly on the status of muscle function, neck, shoulder and back muscles is most tired when maintain 60 degree in normal mattress and most comfortable when 0 degree, but muscles is most comfortable at 30 degree in air mattress. At the same time air mattress is more comfortable than normal mattress especially at 30 and 60 degree.
{"title":"Study on Muscle Fatigue of Human Body Based on Surface Electromyography","authors":"Jianguo Zhang, Haiyan Song, Q. Xue, Taoran Liu, Fang Wang","doi":"10.1109/BMEI.2009.5305065","DOIUrl":"https://doi.org/10.1109/BMEI.2009.5305065","url":null,"abstract":"By studying surface electromyography(sEMG) characteristics of human neck, shoulder and back muscles keeping static posture, to explore the precise evaluation index for muscle function of the human body, and the significance of muscle function prediction, diagnosis and rehabilitation evaluation. Method: 9 experiment objects, the sEMG signals of five muscles (neck extensor, upper trapezius, the middle trapezius, posterior deltoid, latissimus dorsi ) at deferent angles and deferent bed mattresses were tested by 8 channel sEMG collected system. The eigenvalue of MPF( Means power frequency)were processed by SPSS 16.0. Conclusion: deferent angles and deferent mattresses influenced greatly on the status of muscle function, neck, shoulder and back muscles is most tired when maintain 60 degree in normal mattress and most comfortable when 0 degree, but muscles is most comfortable at 30 degree in air mattress. At the same time air mattress is more comfortable than normal mattress especially at 30 and 60 degree.","PeriodicalId":6389,"journal":{"name":"2009 2nd International Conference on Biomedical Engineering and Informatics","volume":"664 ","pages":"1-3"},"PeriodicalIF":0.0,"publicationDate":"2009-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91449232","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-10-30DOI: 10.1109/BMEI.2009.5305604
Bin Zhao, G. Xing
Zhao Bin Xing Geng Yan / the General Hospital of Chinese People’s Armed Police Forces Department of orthopedics 69, yongding Road, Haidian district, Beijing 100039, China. Abstract—In the present study, we investigated the synthesis conditions and the cytocompatibility of single phase -tricalciumphosphate ( -TCP) powder in order to find a very strict regulation of synthesis parameters. The precursors through aqueous precipitation methods were synthesized under several different conditions like reaction pH (3.5~9) and temperature (23~80 ) and flowing nitrogen or not and Ca/P ratio (=1.41 or 1.50) of initial reagents. Eight groups of the powders were obtained after the precursor being sintered at 1100 and characterized by X-ray diffraction and four groups of single phase -TCP powders were studied in cytotoxicity assay. The experimental data indicate that single phase and cytocompatible -TCP powder can be obtained under the reaction condition with the low pH values (pH=4~7), high temperature (50~70 ), flowing nitrogen and strict Ca/P ratio of initial reagents (Ca/P=1.50).
{"title":"Synthesis of beta-Tricalciumphosphate Powder through Aqueous Precipitation Method","authors":"Bin Zhao, G. Xing","doi":"10.1109/BMEI.2009.5305604","DOIUrl":"https://doi.org/10.1109/BMEI.2009.5305604","url":null,"abstract":"Zhao Bin Xing Geng Yan / the General Hospital of Chinese People’s Armed Police Forces Department of orthopedics 69, yongding Road, Haidian district, Beijing 100039, China. Abstract—In the present study, we investigated the synthesis conditions and the cytocompatibility of single phase -tricalciumphosphate ( -TCP) powder in order to find a very strict regulation of synthesis parameters. The precursors through aqueous precipitation methods were synthesized under several different conditions like reaction pH (3.5~9) and temperature (23~80 ) and flowing nitrogen or not and Ca/P ratio (=1.41 or 1.50) of initial reagents. Eight groups of the powders were obtained after the precursor being sintered at 1100 and characterized by X-ray diffraction and four groups of single phase -TCP powders were studied in cytotoxicity assay. The experimental data indicate that single phase and cytocompatible -TCP powder can be obtained under the reaction condition with the low pH values (pH=4~7), high temperature (50~70 ), flowing nitrogen and strict Ca/P ratio of initial reagents (Ca/P=1.50).","PeriodicalId":6389,"journal":{"name":"2009 2nd International Conference on Biomedical Engineering and Informatics","volume":"466 1","pages":"1-6"},"PeriodicalIF":0.0,"publicationDate":"2009-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83021585","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper explores protein-protein interaction extraction from biomedical literature using Support Vector Machines (SVM). Besides common lexical features, various overlap features and base phrase chunking information are used to improve the performance. Evaluation on the AIMed corpus shows that our feature-based method achieves very encouraging performances of 68.6 and 51.0 in F-measure with 10-fold pairwise cross-validation and 10-fold document-wise cross-validation respectively, which are comparable with other state-of-the-art feature-based methods. Keywords-Protein-Protein Interaction; SVM; Shallow Parsing Information
{"title":"Extracting Protein-Protein Interaction from Biomedical Text Using Additional Shallow Parsing Information","authors":"Huanhuan Yu, Longhua Qian, Guodong Zhou, Qiaoming Zhu","doi":"10.1109/BMEI.2009.5302220","DOIUrl":"https://doi.org/10.1109/BMEI.2009.5302220","url":null,"abstract":"This paper explores protein-protein interaction extraction from biomedical literature using Support Vector Machines (SVM). Besides common lexical features, various overlap features and base phrase chunking information are used to improve the performance. Evaluation on the AIMed corpus shows that our feature-based method achieves very encouraging performances of 68.6 and 51.0 in F-measure with 10-fold pairwise cross-validation and 10-fold document-wise cross-validation respectively, which are comparable with other state-of-the-art feature-based methods. Keywords-Protein-Protein Interaction; SVM; Shallow Parsing Information","PeriodicalId":6389,"journal":{"name":"2009 2nd International Conference on Biomedical Engineering and Informatics","volume":"32 1","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2009-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83729996","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-10-30DOI: 10.1109/BMEI.2009.5305309
Min Chen, Shengyong Chen, Q. Guan
Regional difference is the kernel guidance in the area of image segmentation. In this paper, we present a novel region- based active contour model that could be applied to detect the contour of both nucleolus and cell membrane from the background. Both local and global intensity information are used as the driving forces of the active contour model on the principle of maximum regional difference. The local and global fitting forces ensure that local dissimilarities could be captured and global different areas could be segmented respectively. By combining the advantages of local and global information, the motion of contour is driven by the mixed fitting force, which is composed of the local and global fitting term in energy functional. A strategic weight parameter using the gradient information is introduced to explain how the local and global fitting terms work together as the mixed fitting force. Experimental results show desirable performances of our model in segmenting nucleolus.
{"title":"Hybrid Contour Model for Segmentation of Cell Nucleolus and Membranes","authors":"Min Chen, Shengyong Chen, Q. Guan","doi":"10.1109/BMEI.2009.5305309","DOIUrl":"https://doi.org/10.1109/BMEI.2009.5305309","url":null,"abstract":"Regional difference is the kernel guidance in the area of image segmentation. In this paper, we present a novel region- based active contour model that could be applied to detect the contour of both nucleolus and cell membrane from the background. Both local and global intensity information are used as the driving forces of the active contour model on the principle of maximum regional difference. The local and global fitting forces ensure that local dissimilarities could be captured and global different areas could be segmented respectively. By combining the advantages of local and global information, the motion of contour is driven by the mixed fitting force, which is composed of the local and global fitting term in energy functional. A strategic weight parameter using the gradient information is introduced to explain how the local and global fitting terms work together as the mixed fitting force. Experimental results show desirable performances of our model in segmenting nucleolus.","PeriodicalId":6389,"journal":{"name":"2009 2nd International Conference on Biomedical Engineering and Informatics","volume":"15 1","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2009-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83451665","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-10-30DOI: 10.1109/BMEI.2009.5305872
Tianfu Wang, Z. Hao, Chao-Jie Wan
Based on the magnetic resonance imaging (MRI) data, a 3-dimentional finite element model of knee joint was developed using the reverse modeling technology. The model includes bone, all the four main ligaments, menisci and cartilages with consideration of the transversely isotropic constitutive behavior of menisci and hyperelasticity of ligaments. The biomechanics of tibiofemoral joint under an anterior tibial force and a compression load with anterior force was analyzed at three cases: intact anterior cruciate ligament (ACL), partially injured ACL and totally ruptured ACL. Our main goal was to quantitative analyze the effect of ACL injury on the biomechanics of other knee tissues.
{"title":"The Effect of Anterior Cruciate Ligament Injury on the Biomechanical Behavior of Human Knee Joint","authors":"Tianfu Wang, Z. Hao, Chao-Jie Wan","doi":"10.1109/BMEI.2009.5305872","DOIUrl":"https://doi.org/10.1109/BMEI.2009.5305872","url":null,"abstract":"Based on the magnetic resonance imaging (MRI) data, a 3-dimentional finite element model of knee joint was developed using the reverse modeling technology. The model includes bone, all the four main ligaments, menisci and cartilages with consideration of the transversely isotropic constitutive behavior of menisci and hyperelasticity of ligaments. The biomechanics of tibiofemoral joint under an anterior tibial force and a compression load with anterior force was analyzed at three cases: intact anterior cruciate ligament (ACL), partially injured ACL and totally ruptured ACL. Our main goal was to quantitative analyze the effect of ACL injury on the biomechanics of other knee tissues.","PeriodicalId":6389,"journal":{"name":"2009 2nd International Conference on Biomedical Engineering and Informatics","volume":"38 1","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2009-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83562717","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In vivo, cardiomyocytes are stimulated by electricity continuously. So electrical stimulation is an indispensable factor for cardiomyocytes cultured in vitro. In this study, the experimental group neonatal cardiomyocytes received electrical stimulation at different intensities and different frequencies by a electrical stimulation device in vitro. After stimulation, The morphology and beating of cardiomyocytes were observed, and lactate dehydrogenase released in the culture medium, positive ratio of propidium iodide staining, and ATPase activity of the cardiomyocytes were measured. When the electrical stimulation at low intensity and low frequency, the morphology and beating, lactate dehydrogenase released, positive ratio of propidium iodide staining, ATPase activity were all almost unchangeable. However, at higher intensity (10 V/cm) or higher frequency (3.5 Hz), cardiomyocytes stopped beating, lactate dehydrogenase leaking, positive ratio of propidium iodide staining were increased, and ATPase activity was reduced. In conclusion, electrical stimulation at high intensity or frequency damnified neonatal cardiomyocytes in vitro, the proper parameter of electrical stimulation favourable for cardiomyocytes metabolism in vitro was less than 10 V/cm (intensity) and 3.5 Hz (frequency).
{"title":"Effects of Electrical Stimulation on Growth and Metabolism of Cardiomyocytes In Vitro","authors":"Yong Guo, Yan Wei, Xi-zheng Zhang, Yu-xian Yan, Ruixin Li, Chun-qiu Zhang","doi":"10.1109/BMEI.2009.5305424","DOIUrl":"https://doi.org/10.1109/BMEI.2009.5305424","url":null,"abstract":"In vivo, cardiomyocytes are stimulated by electricity continuously. So electrical stimulation is an indispensable factor for cardiomyocytes cultured in vitro. In this study, the experimental group neonatal cardiomyocytes received electrical stimulation at different intensities and different frequencies by a electrical stimulation device in vitro. After stimulation, The morphology and beating of cardiomyocytes were observed, and lactate dehydrogenase released in the culture medium, positive ratio of propidium iodide staining, and ATPase activity of the cardiomyocytes were measured. When the electrical stimulation at low intensity and low frequency, the morphology and beating, lactate dehydrogenase released, positive ratio of propidium iodide staining, ATPase activity were all almost unchangeable. However, at higher intensity (10 V/cm) or higher frequency (3.5 Hz), cardiomyocytes stopped beating, lactate dehydrogenase leaking, positive ratio of propidium iodide staining were increased, and ATPase activity was reduced. In conclusion, electrical stimulation at high intensity or frequency damnified neonatal cardiomyocytes in vitro, the proper parameter of electrical stimulation favourable for cardiomyocytes metabolism in vitro was less than 10 V/cm (intensity) and 3.5 Hz (frequency).","PeriodicalId":6389,"journal":{"name":"2009 2nd International Conference on Biomedical Engineering and Informatics","volume":"15 1","pages":"1-4"},"PeriodicalIF":0.0,"publicationDate":"2009-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80811827","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-10-30DOI: 10.1109/BMEI.2009.5304940
Jinyang Shi, K. Lam, M. Gu, Mingze Li, S. Chung
Wireless body sensor network (WBSN) has gained significant interests as an important infrastructure for the real- time biomedical healthcare system, while the security of the sensitive health information becomes one of the main challenges. Due to the crucial constraints of low power in the sensors, traditional security mechanisms and key distribution schemes are not suitable for WBSN. In this paper, we investigate a novel energy-efficient approach, BodyKey, that can generate and distribute the cryptographic keys using the Electrocardiograph (ECG) biometrics in WBSN. The proposed BodyKey represents the biometric features as ordered set, and deals with the biometric variations using a new fuzzy construction with set reconciliation. In this way, only limited necessary information needs to be communicated for key agreement, and the total energy consumption for key distribution can thus be reduced. The proposed approach has been implemented and evaluated on the public domain PTB Diagnostic Set database with 290 subjects. Experimental results and security analysis for accuracy performance, energy consumption and security property show that BodyKey is a promising and practical key distribution scheme for secure communications in WBSN.
{"title":"Towards Energy-Efficient Secure Communications Using Biometric Key Distribution in Wireless Biomedical Healthcare Networks","authors":"Jinyang Shi, K. Lam, M. Gu, Mingze Li, S. Chung","doi":"10.1109/BMEI.2009.5304940","DOIUrl":"https://doi.org/10.1109/BMEI.2009.5304940","url":null,"abstract":"Wireless body sensor network (WBSN) has gained significant interests as an important infrastructure for the real- time biomedical healthcare system, while the security of the sensitive health information becomes one of the main challenges. Due to the crucial constraints of low power in the sensors, traditional security mechanisms and key distribution schemes are not suitable for WBSN. In this paper, we investigate a novel energy-efficient approach, BodyKey, that can generate and distribute the cryptographic keys using the Electrocardiograph (ECG) biometrics in WBSN. The proposed BodyKey represents the biometric features as ordered set, and deals with the biometric variations using a new fuzzy construction with set reconciliation. In this way, only limited necessary information needs to be communicated for key agreement, and the total energy consumption for key distribution can thus be reduced. The proposed approach has been implemented and evaluated on the public domain PTB Diagnostic Set database with 290 subjects. Experimental results and security analysis for accuracy performance, energy consumption and security property show that BodyKey is a promising and practical key distribution scheme for secure communications in WBSN.","PeriodicalId":6389,"journal":{"name":"2009 2nd International Conference on Biomedical Engineering and Informatics","volume":"114 1","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2009-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88690672","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-10-30DOI: 10.1109/BMEI.2009.5305006
E. Cheng, S. Challa, R. Chakravorty
Automatic cell segmentation in phase contrast mi- croscopy images play a very important role in the study the be- havior of lymphocytes, such as cell motility, cell deformation, and cell population dynamics etc. In this paper, we have developed a set of algorithms for the microscopy image cell segmentation, in which three pairs of edge detection (Sobel, Prewitt and Laplace) based cell segmentation algorithms are developed in parallel to increase the probability of cell detection. Then, an hierarchical model is proposed and used in decision fusion that combine the three pair of detection results to increase the probability of final cell detection. After that, a false removal algorithm is proposed to remove false detections that may occur in the fusion process. The distance and watershed transforms have also been used to separate the connected cells. Experimental results have proved that these algorithms are pretty robust to variable microscopy image data, and variable cell densities, and with the proposed fusion and false removal algorithms, the cell detection rate has increased significantly to above 97% with the false detection rate about 7%. Index Terms—Microscopy Image, Cell Segmentation, Decision Fusion, Edge Detection, Distance Transform, Watershed Trans- form.
{"title":"Microscopic Cell Detection Based on Multiple Cell Image Segmentations and Fusion Algorithms","authors":"E. Cheng, S. Challa, R. Chakravorty","doi":"10.1109/BMEI.2009.5305006","DOIUrl":"https://doi.org/10.1109/BMEI.2009.5305006","url":null,"abstract":"Automatic cell segmentation in phase contrast mi- croscopy images play a very important role in the study the be- havior of lymphocytes, such as cell motility, cell deformation, and cell population dynamics etc. In this paper, we have developed a set of algorithms for the microscopy image cell segmentation, in which three pairs of edge detection (Sobel, Prewitt and Laplace) based cell segmentation algorithms are developed in parallel to increase the probability of cell detection. Then, an hierarchical model is proposed and used in decision fusion that combine the three pair of detection results to increase the probability of final cell detection. After that, a false removal algorithm is proposed to remove false detections that may occur in the fusion process. The distance and watershed transforms have also been used to separate the connected cells. Experimental results have proved that these algorithms are pretty robust to variable microscopy image data, and variable cell densities, and with the proposed fusion and false removal algorithms, the cell detection rate has increased significantly to above 97% with the false detection rate about 7%. Index Terms—Microscopy Image, Cell Segmentation, Decision Fusion, Edge Detection, Distance Transform, Watershed Trans- form.","PeriodicalId":6389,"journal":{"name":"2009 2nd International Conference on Biomedical Engineering and Informatics","volume":"122 1","pages":"1-7"},"PeriodicalIF":0.0,"publicationDate":"2009-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89396511","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：