基于人工神经网络数据挖掘在群体药代动力学中的知识发现

Knowledge discovery based on artificial neural network data mining in population pharmacokinetics

目的人工神经网络(ANN)是人工智能的一个组成部分,本文针对医药信息数据挖掘技术,探讨和总结基于人工神经网络在群体药物动力学中的体系结构。方法对网络数据库中群体药物动力学中的信息样本进行数据挖掘,通过仿真模拟生成大样本群体药代动力学(PPK)数据库,并建立基于SPSS系统的人工神经网络PPK模型。结果模拟的大样本PPK数据与文献报道具有一致的统计学特征,ANN模型建模简单,分析精度良好。结论通过PKK数据挖据及模拟可以重建ANN分析模型。

Objective Artificial neural networks （ANN） is an integral part of artificial intelligence. In this paper, the architecture of artificial neural networks in population pharmacokinetics is discussed and summarized according to the data mining techniques of medical information. Methods The data samples of population pharmacokinetics in network databases are collected by data mining, and a large sample population pharmacokinetics （PPK） database is generated by simulation, and an artificial neural network PPK model based on SPSS system is established. Results The simulated large sample PPK data have the same statistical characteristics as those reported in the literature. The ANN model is simple and the analysis accuracy is better. Conclusion ANN analysis model can be reconstructed by PKK data mining and simulation.