摘要
在对肝脏灌注经典的算法进行比较分析的基础上,改进了灌注图像数据的测量,使用Bayes和SVM分类器进行计算机辅助肝脏病变检测。实验结果表明该文设计的针对MRI灌注图像利用多核SVM算法可以在降低算法复杂度的基础上有效降低分类错误率,从而可以有效进行对肝脏病变的计算机辅助检测。
Computer aided liver perfusion and auxiliary lesion detection technology is an important adjunct to the treatment of liver diseases. Based on comparing and analysising different classical liver perfusion algorithms, this paper uses Bayes and SVM classifier for computer-aided detection of liver lesions. The experimental results show that this algorithm can effectively reduce the classification error rate based on reducing the complexity of algorithm.
出处
《西北大学学报(自然科学版)》
CAS
CSCD
北大核心
2014年第6期904-908,共5页
Journal of Northwest University(Natural Science Edition)
基金
榆林市科技局产学研合作基金资助项目(2012cxy3-5)
西北大学科学研究基金资助项目(ND14042)