多特征向量脑中风微波检测算法探究与改进
|
摘要
以微波成像作为主要手段的脑中风检测技术是一种安全便捷且成本低的微波检测方法,但由于脑部结构复杂,直接成像精度不高效果不可靠。通常采用微波提取脑部特征向量,通过分类模型处理检测脑中风的发生。对适用于多特征向量的脑中风微波检测分类模型进行研究,在构建Adaboost分类器模型基础上,采用分布式粒子群算法对模型进行优化。实验结果表明,将优化后的分类模型用于脑中风检测,精准度高达99.50%,运行速度提升46%,使得脑中风检测精准度和效率得到极大改善。
The microwave detection technology of stroke is a safe, convenient and low-cost shortwave detection method, and mainly uses the microwave imaging technology( MIT). However, due to the complex brain structure, the direct imaging accuracy is not high and the effect is not reliable. It usually extracts the microwave eigenvector of brain and uses the classification model to detect the presence of stroke. This paper makes an exploration on the multi-eigenvector microwave detection model of the brain stroke. It constructs an adaptive boosting(Adaboost) model, and uses the distributed particle swarm optimization(DPSO) to optimize it. The experimental results show that the optimized classification model has an accuracy of 99.50% for stroke detection, and it improves operating speed by 46%, which greatly improves the accuracy and efficiency of stroke detection.
The microwave detection technology of stroke is a safe, convenient and low-cost shortwave detection method, and mainly uses the microwave imaging technology( MIT). However, due to the complex brain structure, the direct imaging accuracy is not high and the effect is not reliable. It usually extracts the microwave eigenvector of brain and uses the classification model to detect the presence of stroke. This paper makes an exploration on the multi-eigenvector microwave detection model of the brain stroke. It constructs an adaptive boosting(Adaboost) model, and uses the distributed particle swarm optimization(DPSO) to optimize it. The experimental results show that the optimized classification model has an accuracy of 99.50% for stroke detection, and it improves operating speed by 46%, which greatly improves the accuracy and efficiency of stroke detection.
引文
[1] Kurz M W, Kurz K D, Farbu E. Acute ischemic stroke-from symptom recognition to thrombolysis[J]. Acta Neurologica Scan- dinavica, 2013, 127(s196): 57-64.
[2] Naghavi M, Wang H, Lozano R, et al. Global, regional, and national age-sex specifi c all-cause and cause-specific mortality for 240 causes of death, 1990-2013: a systematic analysis for the global burden of disease study 2013[C]// Proceedings of the Third International Conference on System Science in Health Care. Springer, 2015: 957-979.
[3] Mustafa S, Mohammed B, Abbosh A. Novel preprocessing techniques for accurate microwave imaging of human brain. IEEE Antennas Wirel Propag Lett, 2013, 12(2): 460-463.
[4] 周宇艳. 基于FDTD的微波脑中风检测算法研究[D].上海:东华大学,2016
[5] Wu Yizhi, Mao Yubo, Saba. The experimental confocal MI system basing on SFCW [C]//2015 International Conference on Information Science and Intelligent Control ( ISIC2015),San-ya,China,2015.
[6] Wu Yizhi, Zhou Yuyan, Wang Yifan, et al. High-resolution microwave stroke detection system based on signal similarity algorithm [C]// 2015 Asia-Pacific Microwave Conference( APMC), Nanjing, 2015.
[7] 魏仕轩,王未央.SVM和集成学习算法的改进和实现[J].计算机系统应用.2015(7):117-121.
[8] 杨新武,马壮,袁顺.基于弱分类器调整的多分类Adaboost算法[J].电子与信息学报,2016,38(2):373-380.
[9] 吴怡之,席恋.基于支持向量机分类的脑中风微波检测[J].微型机与应用,2016,35(13):55-56.
[10] Abbosh, Amin. Microwave-based system using directional wideband antennas for head imaging[C]// International Workshop on Antenna Technology: small Antennas, Novel Em Structures and Materials, and Applications IEEE. Sydney, Australia on March 4-6, 2014:292-295.
[11] 陈雪平,林金官,王晓迪.基于ANOVA高维模型的正交设计的优良性(英文)[J].应用概率统计,2013,29(6):570-580.
[12] Changshui Zhang,Qutang Cai,Yangqiu Song. Boosting with pairwise constraints[J]. Neurocomputing,2009,73(4):908-919.
[13] Xiu Ling Ding, Xue Guang Yuan, Jin Nan Zhang, et al. Adaptive Polarization Mode Dispersion Compensation with DPSO Algorithm[J]. Applied Mechanics and Materials, 2014,3634(687):4015-4018.
[14] 刘登品. P～3的示性函数和Moment-Angel流形的Partial-商[D].上海:复旦大学, 2012.
[15] Ran Li, Guang-Xing Tan, Sheng-Hua Ning. The Research of Multi-pose Face Detection Based on Skin Color and Adaboost [M]. Berlin: Springer, 2011.
[16] 王荣芝,陈晓.基于DPSO算法的并行测试任务调度[J].中国测试,2014,40(3):101-104.
[17] 钟强晖,张志华,梁胜杰.基于比例失效率模型的退化数据分析方法[J].北京航空航天大学学报,2012,38(1):43-47.
[18] 田银, 李沛洋, 徐鹏. 基于AdaBoost的脑机接口分类算法研究[J]. 电子科技大学学报, 2013, 42(5):791-793.
[19] 张涛, 陈万忠, 李明阳. 基于AdaBoost算法的癫痫脑电信号识别[J]. 物理学报, 2015, 64(12):419-425.
[2] Naghavi M, Wang H, Lozano R, et al. Global, regional, and national age-sex specifi c all-cause and cause-specific mortality for 240 causes of death, 1990-2013: a systematic analysis for the global burden of disease study 2013[C]// Proceedings of the Third International Conference on System Science in Health Care. Springer, 2015: 957-979.
[3] Mustafa S, Mohammed B, Abbosh A. Novel preprocessing techniques for accurate microwave imaging of human brain. IEEE Antennas Wirel Propag Lett, 2013, 12(2): 460-463.
[4] 周宇艳. 基于FDTD的微波脑中风检测算法研究[D].上海:东华大学,2016
[5] Wu Yizhi, Mao Yubo, Saba. The experimental confocal MI system basing on SFCW [C]//2015 International Conference on Information Science and Intelligent Control ( ISIC2015),San-ya,China,2015.
[6] Wu Yizhi, Zhou Yuyan, Wang Yifan, et al. High-resolution microwave stroke detection system based on signal similarity algorithm [C]// 2015 Asia-Pacific Microwave Conference( APMC), Nanjing, 2015.
[7] 魏仕轩,王未央.SVM和集成学习算法的改进和实现[J].计算机系统应用.2015(7):117-121.
[8] 杨新武,马壮,袁顺.基于弱分类器调整的多分类Adaboost算法[J].电子与信息学报,2016,38(2):373-380.
[9] 吴怡之,席恋.基于支持向量机分类的脑中风微波检测[J].微型机与应用,2016,35(13):55-56.
[10] Abbosh, Amin. Microwave-based system using directional wideband antennas for head imaging[C]// International Workshop on Antenna Technology: small Antennas, Novel Em Structures and Materials, and Applications IEEE. Sydney, Australia on March 4-6, 2014:292-295.
[11] 陈雪平,林金官,王晓迪.基于ANOVA高维模型的正交设计的优良性(英文)[J].应用概率统计,2013,29(6):570-580.
[12] Changshui Zhang,Qutang Cai,Yangqiu Song. Boosting with pairwise constraints[J]. Neurocomputing,2009,73(4):908-919.
[13] Xiu Ling Ding, Xue Guang Yuan, Jin Nan Zhang, et al. Adaptive Polarization Mode Dispersion Compensation with DPSO Algorithm[J]. Applied Mechanics and Materials, 2014,3634(687):4015-4018.
[14] 刘登品. P～3的示性函数和Moment-Angel流形的Partial-商[D].上海:复旦大学, 2012.
[15] Ran Li, Guang-Xing Tan, Sheng-Hua Ning. The Research of Multi-pose Face Detection Based on Skin Color and Adaboost [M]. Berlin: Springer, 2011.
[16] 王荣芝,陈晓.基于DPSO算法的并行测试任务调度[J].中国测试,2014,40(3):101-104.
[17] 钟强晖,张志华,梁胜杰.基于比例失效率模型的退化数据分析方法[J].北京航空航天大学学报,2012,38(1):43-47.
[18] 田银, 李沛洋, 徐鹏. 基于AdaBoost的脑机接口分类算法研究[J]. 电子科技大学学报, 2013, 42(5):791-793.
[19] 张涛, 陈万忠, 李明阳. 基于AdaBoost算法的癫痫脑电信号识别[J]. 物理学报, 2015, 64(12):419-425.