Silva Alvaro, Cortez Paulo, Santos Manuel Filipe, Gomes Lopes, Neves José
Serviço de Cuidados Intensivos, Hospital Geral de Santo António, Porto, Portugal.
Artif Intell Med. 2006 Mar;36(3):223-34. doi: 10.1016/j.artmed.2005.07.006. Epub 2005 Oct 6.
This work presents a novel approach for the prediction of mortality in intensive care units (ICUs) based on the use of adverse events, which are defined from four bedside alarms, and artificial neural networks (ANNs). This approach is compared with two logistic regression (LR) models: the prognostic model used in most of the European ICUs, based on the simplified acute physiology score (SAPS II), and a LR that uses the same input variables of the ANN model.
A large dataset was considered, encompassing forty two ICUs of nine European countries. The recorded features of each patient include the final outcome, the case mix (e.g. age) and the intermediate outcomes, defined as the daily averages of the out of range values of four biometrics (e.g. heart rate). The SAPS II score requires 17 static variables (e.g. serum sodium), which are collected within the first day of the patient's admission. A nonlinear least squares method was used to calibrate the LR models while the ANNs are made up of multilayer perceptrons trained by the RPROP algorithm. A total of 13,164 adult patients were randomly divided into training (66%) and test (33%) sets. The two methods were evaluated in terms of receiver operator characteristic (ROC) curves.
The event based models predicted the outcome more accurately than the currently used SAPS II model (P<0.05), with ROC areas within the ranges 83.9-87.1% (ANN) and 82.6-85.2% (LR) versus 80% (LR SAPS II). When using the same inputs, the ANNs outperform the LR (improvement of 1.3-2%).
Better prognostic models can be achieved by adopting low cost and real-time intermediate outcomes rather than static data.
本研究提出了一种基于不良事件预测重症监护病房（ICU）死亡率的新方法，不良事件由四个床边警报定义，同时结合人工神经网络（ANN）。该方法与两种逻辑回归（LR）模型进行比较：大多数欧洲ICU使用的基于简化急性生理学评分（SAPS II）的预后模型，以及使用与ANN模型相同输入变量的LR模型。
考虑了一个大型数据集，涵盖九个欧洲国家的42个ICU。每个患者记录的特征包括最终结局、病例组合（如年龄）和中间结局，中间结局定义为四种生物特征（如心率）超出范围值的每日平均值。SAPS II评分需要17个静态变量（如血清钠），这些变量在患者入院第一天内收集。使用非线性最小二乘法校准LR模型，而ANN由通过RPROP算法训练的多层感知器组成。总共13164名成年患者被随机分为训练集（66%）和测试集（33%）。通过受试者操作特征（ROC）曲线对这两种方法进行评估。
基于事件的模型比目前使用的SAPS II模型更准确地预测结局（P<0.05），ROC面积范围为83.9 - 87.1%（ANN）和82.6 - 85.2%（LR），而LR SAPS II为80%。当使用相同输入时，ANN的表现优于LR（提高1.3 - 2%）。
采用低成本的实时中间结局而非静态数据可实现更好的预后模型。
