Learning Polynomial Networks for Classification of Clinical Electroencephalograms

TL;DR

This paper introduces a polynomial network approach using evolutionary strategies for classifying noisy clinical EEG data, including Alzheimer detection and artifact recognition, producing interpretable models.

Contribution

The paper presents a novel polynomial network method with evolutionary learning for EEG classification, emphasizing model interpretability and robustness to noise.

Findings

Effective classification of EEGs from Alzheimer and healthy patients.

Models are easy for experts to interpret.

Competitive performance compared to other machine learning methods.

Abstract

We describe a polynomial network technique developed for learning to classify clinical electroencephalograms (EEGs) presented by noisy features. Using an evolutionary strategy implemented within Group Method of Data Handling, we learn classification models which are comprehensively described by sets of short-term polynomials. The polynomial models were learnt to classify the EEGs recorded from Alzheimer and healthy patients and recognize the EEG artifacts. Comparing the performances of our technique and some machine learning methods we conclude that our technique can learn well-suited polynomial models which experts can find easy-to-understand.