The Combined Technique for Detection of Artifacts in Clinical Electroencephalograms of Sleeping Newborns

TL;DR

This paper introduces a hybrid method combining polynomial neural networks and decision trees to effectively detect artifacts in sleep EEGs of newborns, improving accuracy over existing techniques.

Contribution

The paper presents a novel combined approach that enhances artifact detection in neonatal sleep EEGs by generating interpretable classification rules with lower error.

Findings

Outperforms standard machine learning methods in artifact detection accuracy

Produces comprehensible classification rules for clinical interpretation

Effectively handles heavily corrupted EEG data

Abstract

In this paper we describe a new method combining the polynomial neural network and decision tree techniques in order to derive comprehensible classification rules from clinical electroencephalograms (EEGs) recorded from sleeping newborns. These EEGs are heavily corrupted by cardiac, eye movement, muscle and noise artifacts and as a consequence some EEG features are irrelevant to classification problems. Combining the polynomial network and decision tree techniques, we discover comprehensible classification rules whilst also attempting to keep their classification error down. This technique is shown to outperform a number of commonly used machine learning technique applied to automatically recognize artifacts in the sleep EEGs.