Energy landscapes for a machine-learning prediction of patient discharge
Energy landscapes for a machine-learning prediction of patient discharge
复制标题
DOI：
10.1103/physreve.93.063310
复制
发表时间：
2016-06-17
影响因子：
2.4
通讯作者：
Wales, David J.
中科院分区：
文献类型：
--
作者：
Das, Ritankar;Wales, David J.
能源景观框架应用于通过训练神经网络参数生成的配置空间。在这项研究中，输入数据包括对医院患者监测的生命体征集合的时间序列，结果是患者出院或继续住院。使用机器学习作为预测诊断工具来实时识别大量电子健康记录数据中的模式是支持临床决策的一种非常有吸引力的方法，它有可能改善患者的治疗结果并减少出院的等待时间。在这里，我们报告了一些初步分析，以展示如何应用机器学习。特别是，我们根据局部最优神经网络以及参数空间中它们之间的连接来可视化拟合景观。我们预计这些结果以及分子系统热力学性质的类似物可能有助于未来设计改进的预测工具。
The energy landscapes framework is applied to a configuration space generated by training the parameters of a neural network. In this study the input data consists of time series for a collection of vital signs monitored for hospital patients, and the outcomes are patient discharge or continued hospitalisation. Using machine learning as a predictive diagnostic tool to identify patterns in large quantities of electronic health record data in real time is a very attractive approach for supporting clinical decisions, which have the potential to improve patient outcomes and reduce waiting times for discharge. Here we report some preliminary analysis to show how machine learning might be applied. In particular, we visualize the fitting landscape in terms of locally optimal neural networks and the connections between them in parameter space. We anticipate that these results, and analogues of thermodynamic properties for molecular systems, may help in the future design of improved predictive tools.