Analysis of EEG rhythms using custom-made MatLab application for processing of data collected during neurofeedback training in ADHD subjects.

EEG;ADHD;Signal processing;Neurofeedback

Brent Hillard

University:University of Louisville

Department:Bioengineering

关键词: EEG;    ADHD;    Signal processing;    Neurofeedback;   

Others  :  https://ir.library.louisville.edu/cgi/viewcontent.cgi?article=1619&context=etd

美国|英语

来源: The Universite of Louisville's Institutional Repository

【 摘 要 】

INTRODUCTION: Attention Deficit/Hyperactivity Disorder (ADHD) is a disorder that is prevalent throughout the world. It is believed that 5% of school aged children suffer from ADHD, with some estimates indicating as high as 10% may suffer from the disorder. Primarily, three subtypes of ADHD have been associated with certain electroencephalographic (EEG) abnormalities. The most common treatment for ADHD is medication. However, neurofeedback is considered an efficacious treatment for ADHD. It is proposed that neurofeedback training aimed to mitigate inattention and low arousal in ADHD will be accompanied by changes in EEG bands' relative power. DATA COLLECTION: Patients were 18 children with ADHD. The neurofeedback protocol used to train patients has focused attention training mode, which according to specifications, represents an EEG desynchronization measure. The neurofeedback protocol provides one numeric output for "Focus” and one numeric output for “Alertness”. This does not allow for collecting information regarding changes of specific EEG bands’ power within 2-45 Hz range. Therefore, data was collected for EEG bands through the use of BioExplorer and BioReview software. Each subject completed 12 sessions with a target length of 25 minutes per session. Additionally, IVA+Plus test and ABC behavioral survey measures were administered both pre- and post- neurofeedback. DATA PROCESSING AND ANALYSIS: Analysis was completed on each of the 25 min long twelve sessions using a custom-made MatLab application to determine the relative power of each of the EEG bands throughout each session and from the first session to the last session. Additional statistical analysis was performed to determine significant changes in relative power within sessions (from minute 1 to minute 25), and between sessions (from session 1 to session 12) for an individual patient using an ANOVA. Additionally, a correlation analysis was performed to determine possible correlations between the “Focus” measure and changes in relative power of Theta, Alpha, Low and High Beta, Theta/Alpha, Theta/Beta, and Theta/Low Beta and Theta/High Beta ratios. Additional measures of patients’ post-neurofeedback outcomes were assessed using an audio-visual selective attention test and behavioral evaluation scores and analyzed through a paired t-test. RESULTS: Through statistical analysis of the data computed in the MatLab application, we determined that, as expected, Theta/Low Beta and Theta/Alpha ratios decreased significantly from Session 1 to Session 12 and from minute 1 to minute 25 within sessions. "Focus" measure also demonstrated a significant gradual increase from session 1 to session 12 and from minute 1 to minute 25 within