[SP4-3B-3] A new QEEG Method for Better Diagnosis of ADHD by Time-domain Analysis
Attention-deficit/hyperactivity disorder (ADHD) is one of the most common neurodevelopmental disorders of childhood. The diagnosis of ADHD is based on the core symptoms appearing in DSM-V or checklists. However, either symptoms or checklists are subjective, which leads to the problem of overdiagnosis or underdiagnosis. Therefore, many studies have used electro-encephalography (EEG) as a biomarker, in children with ADHD compared with normal controls to assist ADHD diagnosis. Theta/beta ratio (TBR) was approved by FDA as a tool to aid in the diagnosis of ADHD. However, in recent years, this finding has not been replicated in at least five studies with low sensitivity and specificity. In this study, we try to use QEEG analysis as a biomarker to assist ADHD diagnosis by Time-domain Analysis. Since the male prevalence of ADHD, we decided to include male subjects only. 25 ADHD boys and 23 age, sex matched control were enrolled. In this study, a classification analysis-based approach composed of a training phase and a classification phase was developed for classifying the subject’s EEG features as ADHD or normal. There were eight features selected for classification. The classification results showed significantly higher TAR_CZ_std, RelPowGamma_FP2_std, RelPowGamma_FP1_avg, DGR_CZ_snr, RelPowGamma_PZ_snr, Wavelet_db4_EnergyBand_1_T6_avg, and SpectrEdgeFreq_C4_snr in the ADHD group as compared to the normal subjects. On the contrary, there was significantly lower RelPowAlpha_T3_snr in the ADHD group as compared to the normal subjects. The analyses yielded a precision rate, sensitivity and specificity of 92.67%, 90.00%, 96.67%, respectively. Therefore, the developed method is a useful tool in identifying the patients with ADHD precisely.