Constructing a diagnostic-prediction model for Attention-Deficit Hyperactivity Disorder
Approved Research ID: 85636
Approval date: August 22nd 2022
Attention-Deficit/Hyperactivity Disorder (ADHD) is a neurodevelopmental disorder with a prevalence of about 2.5% in adults and 5% in school-aged children. Twin studies have shown that the disorder is inherited with a probability of about 70%. The current diagnostic criteria for ADHD are mostly subjective, and the diagnosis is made without the use of genetic biomarkers, even though previous studies have strongly suggested that the disorder is hereditary. The results of the genome-wide association study did not identify any genes with large effect sizes that are closely related to the expression mechanism of ADHD, as reported in previous studies. This suggests that many genetic factors with small effects are involved in ADHD, and that the phenotype of ADHD is not manifested only by specific genes. Therefore, we are conducting research activities to construct a diagnostic prediction model of ADHD based on genome-wide genetic information by utilizing machine learning and other methods with many candidate genetic causes reported so far to be associated with ADHD. The purpose of this study is to overcome the problem that ADHD has ambiguous criteria for diagnosis, and to construct a predictive model that assists the diagnosis of ADHD with objective indicators. The project period is 24 months.
As described in A6, it is expected that the diagnosis of psychiatric disorders, including ADHD, can be made objectively using genetic and biological indicators.