论文部分内容阅读
目的 利用两个预后分析模型———time -codedmodel和single -timepointmodel ,结合实际数据进行肺癌预后结果预测。方法 利用LM算法和停止准则进行迭代。结果 利用BP人工神经网络的两个模型对于学习集数据的拟合情况要远远好于传统的Cox回归和logistic回归 ,若样本收集具有代表性 ,该网络可以任意精度逼近任意映射。利用停止准则后 ,在数据收集不是非常全面的条件下 ,预测效果与传统方法无差别。结论 BP人工神经网络可以用于肺癌预后结果预测。
Objective To use two prognostic analysis models—time-codedmodel and single-timepointmodel, combined with actual data to predict the prognosis of lung cancer. Method Iterates using the LM algorithm and stop criteria. Results The two models using BP artificial neural network are much better than the traditional Cox regression and logistic regression in fitting the learning set data. If the sample collection is representative, the network can approximate any mapping with arbitrary precision. After using the stopping criterion, the forecasting effect is no different from the traditional method under the condition that the data collection is not very comprehensive. Conclusion BP artificial neural network can be used to predict the prognosis of lung cancer.