论文部分内容阅读
目的建立一种计算机辅助诊断(CAD)模型对系统性红斑狼疮(SLE)合并肾受累进行分类辅助诊断,为及时发现并确诊该疾病提供一种新的方法。方法利用模拟退火(SA)算法优化支持向量机(SVM)算法惩罚系数C和核函数参数g,并基于此构建SA-SVM分类器模型,将其应用于SLE的智能辅助诊断。结果该方法避免了单一SVM分类器易陷入局部最优的缺点,提高了分类器的分类精度,且针对SLE合并肾受累疾病的分类准确率最高可达98.72%。结论实验结果表明该分类模型在SLE合并肾受累疾病智能诊断中有较好的应用价值。
Objective To establish a computer aided diagnosis (CAD) model to classify the diagnosis of systemic lupus erythematosus (SLE) with renal involvement and provide a new method for timely detection and diagnosis of the disease. Methods The SA algorithm was used to optimize the SVM penalty coefficient C and the kernel function g. Based on this, the SA-SVM classifier model was constructed, which was applied to the intelligent diagnosis of SLE. Results The proposed method avoids the disadvantage of single SVM classifier falling into the local optimum and improves the classification accuracy of the classifier. The classification accuracy of the model for SLE with renal involvement is up to 98.72%. Conclusion The experimental results show that the classification model has a good value in the intelligent diagnosis of SLE combined with renal involvement disease.