Sleep Stage Classification: Scalability Evaluations of Distributed Approaches

TL;DR

This paper evaluates the scalability of distributed machine learning algorithms for classifying sleep stages from EEG data using a big data framework built on SparkMLlib, aiming to improve processing efficiency of large clinical datasets.

Contribution

It introduces a scalable big data framework for sleep stage classification from EEG signals and assesses the scalability of various classification algorithms on large datasets.

Findings

Distributed algorithms show promising scalability for EEG sleep data

SparkMLlib effectively handles large-scale sleep disorder diagnosis data

Evaluation results guide future development of scalable clinical data analysis tools

Abstract

Processing and analyzing of massive clinical data are resource intensive and time consuming with traditional analytic tools. Electroencephalogram (EEG) is one of the major technologies in detecting and diagnosing various brain disorders, and produces huge volume big data to process. In this study, we propose a big data framework to diagnose sleep disorders by classifying the sleep stages from EEG signals. The framework is developed with open source SparkMlib Libraries. We also tested and evaluated the proposed framework by measuring the scalabilities of well-known classification algorithms on physionet sleep records.