基于克隆算法的乳腺X线影像计算机辅助诊断算法（英文）

Mammography microcalcification detection based on clonal algorithm

乳腺钼靶X线片微钙化点自动提取是乳腺图像处理中的热点问题。许多算法已经被提出,但由于阳性检出率低,假阳性率高,不能完全满足临床应用的需要。本文提出一种基于克隆算法的微钙化点自动提取方法,目的是获得高阳性检出率和较低的假阳性率。该方法基于对克隆技术分析和假设的基础上,模拟动物克隆过程,建立一种新的克隆算法数学模型。本文数学模型被用于乳腺钼靶X片微钙化点自动提取,取得较好效果。计算机仿真结果验证了本文算法的有效性。

Automatic detection of microcalcifications in breasts has become a hot issue in computer-aided diagnosis. Many detection methods have been put forward, but the positive detection rate and false positive rate of most of methods are unsatisfactory, without fully meeting the clinical needs. A new mathematical model of clonal algorithm was proposed in the paper. Based on the analysis and assumption on cloning technology, a new mathematical model was applied to extract microcalcifications, achieving high positive detection rate and low false positive rate. And the effectiveness of the algorithm has been proved by computer simulation results.