Examining The Effect of Feature Selection on Improving Patient Deterioration Prediction

Abstract

Large amount of heterogeneous medical data is generated every day in various healthcare organizations. Those data could derive insights for improving monitoring and care delivery in the Intensive Care Unit. Conversely, these data presents a challenge in reducing this amount of data without information loss. Dimension reduction is considered the most popular approach for reducing data size and also to reduce noise and redundancies in data. In this paper, we are investigate the effect of the average laboratory test value and number of total laboratory in predicting patient deterioration in the Intensive Care Unit, where we consider laboratory tests as features. Choosing a subset of features would mean choosing the most important lab tests to perform. Thus, our approach uses state-of-the-art feature selection to identify the most discriminative attributes, where we would have a better understanding of patient deterioration problem. If the number of tests can be reduced by identifying the most important tests, then we could also identify the redundant tests. By omitting the redundant tests, observation time could be reduced and early treatment could be provided to avoid the risk. Additionally, unnecessary monetary cost would be avoided. We apply our technique on the publicly available MIMIC-II database and show the effectiveness of the feature selection. We also provide a detailed analysis of the best features identified by our approach.

The impact of extraneous features on the performance of recurrent neural network models in clinical tasks

“…Motivated by lowering patient discomfort and financial costs, AlNuaimi, et. al investigated the effect of reducing the number of lab tests on model performance 12 . Starting with 35 lab variables as the original inputs for modeling patient deterioration, they repeatedly lowered the number of input variables and employed feature selection algorithms to identify the optimal set of variables.…”

Towards human activity recognition for ubiquitous health care using data from awaist-mounted smartphone

“…In the feature selection step, a features subset is selected from the candidate's feature set. There are a large number of studies [27][28][29][30] in which different feature selection techniques have been adapted for supporting the health care, early stroke diagnosis, elderly care, and other medical monitoring applications. In these studies, different techniques such as genetic algorithm [31], sequential forward floating search (SFFS) [32], iterative search margin based algorithm (Simba) [33], minimal redundancy maximal relevance criterion (mRMR) [34], and Relief [35] were used for feature selection.…”

An Efficient Feature Selection Model for IGBO Text

“…The major problem of text-based system is the enormous quantity of features which reduces system performance and consumes higher time [3]. Noura et al defined feature selection, as the process of choosing important features for use in text model construction to improve the performance of the model [4]. Feature selection process is highly recommended in any text-based system to select the most relevant features, thereby reducing the feature dimension space and improving the system performance.…”