基于参保人诊疗经历的住院时长预测模型

首先，通过贵州省医疗保障事务中心（暨我单位）收集整理的数据，并对数据进行脱密脱敏，并结合各医院的就诊规则等相关数据作为训练用数据集；对数据集中的特征和目标变量进行分割，并将特征标准化，把所有特征的数据都调整到相同的“尺度”，一般变成均值为0，标准差为1的形式，使模型更容易学习数据之间的关系。其次，将训练用数据集分为训练集和测试集，使用训练数据集对ANN（Artificial Neural Network）模型进行训练，使用测试集进行测试。最后，使用训练好的ANN模型进行预测操作，模型的输入参数包括但不限于患者姓名、患者年龄、患者性别、医院编号、医院等级、上次住院时间、住院次数、入院病情、就诊病况、使用药物、历史病症等等。First, the training dataset was collected and curated from the data of Guizhou Provincial Medical Insurance Service Center (our unit), with all data undergoing de-identification and desensitization, and combined with relevant data including diagnosis and treatment rules of various hospitals. Subsequently, features and target variables in the dataset were split, and feature standardization was implemented to scale all feature data to a unified scale, generally adjusted to a mean of 0 and standard deviation of 1, to help the model more easily learn the relationships within the data. Second, the training dataset was divided into a training set and a test set. The ANN (Artificial Neural Network) model was trained using the training dataset and evaluated with the test set. Finally, the trained ANN model was used for prediction tasks. The input parameters of the model include, but are not limited to, patient's name, age, gender, hospital ID, hospital level, last hospitalization time, number of hospitalizations, admission condition, current clinical condition, medications used, historical medical conditions, and so on.

搜集汇总

背景与挑战

背景概述

该数据集由贵州省卫生项目管理办公室提供，用于开发住院时长预测模型，数据规模1TB，每月更新。模型通过人工神经网络分析患者诊疗数据，旨在优化患者和医院的日程管理。