Sunday 02 February 2025
A new approach to diagnosing attention-deficit/hyperactivity disorder (ADHD) has been developed, using a combination of electroencephalography (EEG) and deep learning techniques. This method could potentially provide a more accurate and accessible way of identifying children with ADHD.
Currently, ADHD is diagnosed through behavioral observations, which can be subjective and prone to misdiagnosis. The new approach uses EEG signals recorded from children’s brains while they perform simple tasks, such as counting characters on a screen. These signals are converted into spectrograms, which are then analyzed by a deep learning model to identify patterns associated with ADHD.
The study used a publicly available dataset of EEG recordings from 61 children with ADHD and 60 healthy controls aged 7-12. The results showed that the deep learning model was able to accurately classify the children as having ADHD or not, with a high precision, recall, and F1 score of 0.9.
The researchers also found that certain brain regions, including the frontopolar, parietal, and occipital lobes, were significantly affected in children with ADHD. These findings support current research suggesting decreased gray matter in these areas of the brain.
To develop a digital diagnostic system, the researchers used the features extracted from the EEG data to create a targeted cognitive test system. This system consists of three tests that assess different aspects of brain function, including attention, spatial awareness, and memory. The reaction time and accuracy of each test can be collected to provide a more comprehensive assessment of a child’s behavior.
This new approach has the potential to revolutionize the way ADHD is diagnosed, particularly in school settings where early identification and intervention are crucial for optimal development outcomes. The digital diagnostic system could also be used to monitor treatment effectiveness and adjust therapy plans as needed.
The study highlights the importance of using objective measures to diagnose ADHD, rather than relying solely on behavioral observations. By leveraging advances in EEG technology and deep learning algorithms, this new approach offers a more accurate and accessible way to identify children with ADHD, ultimately leading to better outcomes for those affected by the disorder.
Cite this article: “Accurate and Accessible Diagnosis of ADHD Using EEG and Deep Learning Techniques”, The Science Archive, 2025.
Adhd, Eeg, Deep Learning, Diagnosis, Brain Function, Cognitive Test, Attention, Spatial Awareness, Memory, Digital Diagnostic System
