人工神经网络用于消化道癌症初级诊断研究

Artificial Neural Network Applied to Diagnosis of Digestive Tract Cancer

测定了正常人及消化道癌症病人头发样品１１种元素含量，然后将人工神经网络（ＡＮＮ）用于正常人与癌症患者的分类预测，并用独立预测样本作了检验，预报识别率达１００％。讨论了网络参数的选择和发样中微量元素与癌症的关系。结果表明该方法可作为消化道癌症初级诊断的一种辅助手段。

The contents of 11 trace elements in hair sapmles including both mormal people and digestive tract cancer patients were determined by inductively coupled plasma stomic emission spectrometry and fluorescence analysis. Artificial neural network (ANN) approach was applied to classification of the two groups. This method was verified with independent prediction samples. The signification of elements in hair samples for classification and network parameters as also investigaed. The results showed that the artificial neural network model is a good prediction method, therefore, the model might be referred as an aiding means of the diagnosis for the cancer.