基于脑部MR图像GMM特征决策分类的肿瘤诊断

发布时间：2018-01-24 05:27

  本文关键词： 脑部肿瘤 磁共振图像 分割 高斯混合模型特征 决策树　出处：《控制工程》2017年08期 　论文类型：期刊论文

【摘要】：磁共振(MR)图像提供了大量用于医疗检查的信息。精确鲁棒的脑部MR图像分割、特征提取和分类对于临床诊断肿瘤是非常重要的。提出一种新的基于脑部MR图像的肿瘤诊断方法。首先,通过多阈值分割形态学操作检测图像的畸形区域,然后,提取用于分类的高斯混合模型(GMM)特征,最后,利用决策树分类器对肿瘤图像类型进行分类。整个分类过程分为训练和测试2个阶段,训练阶段提取肿瘤图像和非肿瘤图像不同的特征,在测试阶段基于知识库进行肿瘤和非肿瘤分类。使用准确度、误报率和漏检率3个性能指标对算法进行评估,实验结果表明,分类准确度可达91.18%-94.11%,误报率和漏检率在2.94%-4.41%范围内,可以有助于更好的脑部肿瘤诊断。
[Abstract]:Magnetic resonance (MR) images provide a lot of information for medical examination. Brain MR image accurate and robust segmentation, feature extraction and classification is very important for the clinical diagnosis of the tumor. Propose a new tumor diagnosis method based on MR image of the brain. First of all, the deformed regions, threshold segmentation morphology detection image then, to extract the Gauss mixture model classification (GMM) characteristics, finally, the classification of tumor image types by using decision tree classifier. The classification process is divided into 2 stages of training and testing, the extraction of tumor images and non tumor images with different characteristics of the training stage, tumor and non tumor classification in the testing stage based on the use of the knowledge base. The accuracy, false positive rate and false negative rate of 3 indicators to evaluate the performance of the algorithm, the experimental results show that the classification accuracy of 91.18%-94.11%, false alarm rate and false negative rate in 2.94% Within the range of -4.41%, it can help better diagnosis of brain tumors.

【作者单位】： 包头医学院计算机科学与技术系;西南科技大学国防科技学院;
【基金】：内蒙古自治区高等学校科学研究项目(NJZY13252) 包头医学院科学研究基金项目(BYJJ-QM201657)
【分类号】：R730.44;TP391.41
【正文快照】： 1引言 磁共振(MRI)在医疗领域起着至关重要作用,波场不均匀性引起的强度变化。不均匀性,特别是 为医疗诊断提供定性、定量和准确信息。其在许多应用领域比其他模态医疗成像技术更具优势,如在心血管、神经、肌肉疾病诊断,特别是脑成像领域。然而,多数MR图像处理的问题起因于B，

本文编号：1459280