摘要
由于乳腺X光图像的复杂性,直接从图像中看出肿瘤及其良、恶性质是比较困难的,因此建立高效的肿瘤自动诊断系统是非常必要的。文章将粗糙集理论中基于信息增益的约简方法和神经网络相结合,提出了粗糙神经网络算法RNN,将其应用于乳腺X光图像分类。实验结果表明,该方法的分类精确度可达到92.37%比单独使用神经网络方法的分类精确度(81.25%)要高,同时所花费的时间也明显减少。
Detecting tumor in mammography is a difficult task Because of complexity in the image. This brings the necessity of creating automatic tools to find whether a mammography present tumor or not. In this paper we join neural network with information gain reduction of rough sets theory which we call the rough neural network（RNN）to classify digital mammography. The experimental results show that the rough neural network performs better than only neural network algorithm in terms of time though it can get 92.37% classify accuracy which is higher than 81.25% using neural network only.
出处
《计算机科学》
CSCD
北大核心
2006年第11期151-153,共3页
Computer Science
基金
国家自然科学基金资助项目(60373108)
国家自然科学基金资助项目(60573096)
甘肃省自然科学基金资助项目(3ZS051-A25-042)
