Binary particle swarm optimization for feature Selection on uterine electrohysterogram signal

Abstract

The Selection of pertinent features is a very important problem in pattern recognition. Therefore, we need reliable feature selection methods to reduce the number of features, through the elimination of irrelevant and noisy features. In our study we try to detect the pertinent features extracted from uterine electrohysterography that permit to classify at best labor and pregnancy contractions. The global aim of this work is to detect the preterm deliveries. In this paper we present a feature selection method based on binary particle swarm optimization. The performance of this method was tested by using three types of classifiers. The results show that this method gives common features whatever the type of classifiers.

Entity Extraction in Biomedical Corpora: An Approach to Evaluate Word Embedding Features with PSO based Feature Selection

“…Here we propose a PSO based feature selection technique that determines the most relevant features from a full word embedding set, and use this subset as feature for classifier's training. Feature selection has been widely used for many tasks such as gene expression (Ding and Peng, 2005), face recognition (Seal et al, 2015) and signal processing (Alamedine et al, 2013). Dealing with biomedical text is, however, more difficult and challenging as the features have non-numeric values and the texts are heavily unstructured.…”

N on - Invasive Feto - Maternal Well - Being Monitoring: A Review of Methods

“…This work is further extended by J Laforet et al who proposed an electrophysiological model that aims at describing the multiscale evolution of uterine activity from its genesis at the cell level to its propagation at the myometrium level up to its projection through the volume conductor tissue to the abdominal surface (EHG) [136]. D Alamedine et al developed a feature selection method based on Particle Swarm Optimisation (PSO) and the classifiers used has a strong influence on selected EHG features which can give relevant information [137]. K Horoba et al extracted slow waves from EHG signals as they resemble with mechanical signals and can be regarded as contraction waves.…”