Open AccessNovel approach for sleep disorder monitoring using a finite‐state machine for localities lacking specialist physicians
Author(s) -
Swetapadma Aleena
Publication year - 2017
Publication title -
iet science, measurement and technology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.418
H-Index - 49
eISSN - 1751-8830
pISSN - 1751-8822
DOI - 10.1049/iet-smt.2017.0240
Subject(s) - finite state machine , electroencephalography , sleep (system call) , computer science , pattern recognition (psychology) , automaton , fault (geology) , electromyography , electrooculography , artificial intelligence , sleep stages , speech recognition , eye movement , algorithm , polysomnography , psychology , physical medicine and rehabilitation , medicine , neuroscience , seismology , geology , operating system
This study proposes a novel method for sleep disorder monitoring based on a finite‐state machine (FSM) from various bio‐signals, namely electroencephalography (EEG), electro‐oculography (EOG) and electromyography (EMG) signals. The sleep signals have been obtained from physionet sleep repository, which includes horizontal EOG, submental‐EMG and EEG sampled at 100 Hz sampling frequency. Inputs given to the FSM‐based module are the processed signals from EMG, EEG and EOG signals. The FSM module for sleep analysis is composed of different states and the conditions to flow from one state to another state. In this study, two FSM modules are designed, one for sleep wave and another for sleep stage identification. Based on the outputs obtained from the above two FSM modules, the sleep disorder can be monitored. The accuracy of the proposed method has been calculated with percentage accuracy, false acceptance rate and false rejection ratio. The average classification accuracy of the finite‐state automaton (FSA)‐based method is up to 99% for all the tested fault cases. The proposed FSA method suggests a novel method and can be put to effective use in the rural areas for primary analysis.