
This paper image segmentation method is utilized to extract the lung cancer lung parenchyma of CT image, and create its gray level co-occurrence matrix and gray level co-occurrence matrix is utilized to extract image texture feature of lung parenchyma.Then texture feature dataset for SPSS pca dimension reduction, generate new data samples, the use of BP neural network, RBF neural network and support vector machine learning algorithm for training test.The classification results show that the degree of sensitivity and the characteristics of BP neural network classification results can reach above 0.8 and the coincidence rate is relatively high.Support vector machine (SVM) classifier classification results followed, RBF used in the classification of the worst.
Key words: lung cancerGray level co-occurrence matrixTexture featureNeural networkSupport vector machine (SVM)
The pulse signal is the life of the human basic symbol, one of the pulse signal detection in clinical, teaching and scientific research plays a very important role.Current pulse signal monitoring is using special medical instrument, the instrument is fixed, single function, huge volume, high price, analysis of such problems as imperfect.In this paper, a pulse signal acquisition and analysis system was designed and developed.The system mainly consists of two parts of hardware and software.Hardware part using HK - 2000 series integrated pulse sensor complete the pulse signal acquisition and regulate.Software part do the login interface with LabVIEW and main form interface, signal acquisition interface by VB programming, VB program to realize the calling through the LabVIEW real-time display and storage of pulse waveform.This system friendly interface, simple operation, enhance the function of automatic analysis of traditional medical instruments, has a certain application value.
采纳我的我的比楼下好
欢迎分享,转载请注明来源:内存溢出
微信扫一扫
支付宝扫一扫
评论列表(0条)