ADHD (Attention-Deficit/Hyperactivity Disorder) is a complex neurodevelopmental condition affecting millions of individuals worldwide. Traditional diagnostic methods often rely on subjective assessments, leaving room for variability in interpretation. Recent advancements in neuroscience, particularly in EEG (electroencephalography) technology, have opened new pathways for objective analysis. This article delves into the use of EEG to analyze brainwave activity—specifically theta and beta waves—offering a deeper understanding of ADHD's neurobiological underpinnings and presenting innovative approaches for improving diagnostic accuracy and treatment outcomes.
By Abbie Bull, University of Birmingham, in collaboration with Naxon Labs
Attention-Deficit/Hyperactivity Disorder (ADHD) is a complex neurodevelopmental condition characterized by inattention, hyperactivity, and impulsivity. One promising avenue for understanding and diagnosing ADHD lies in electroencephalography (EEG), which provides a window into brain activity patterns. Specifically, the theta/beta ratio (TBR) has emerged as a neural marker of ADHD. Theta waves (4–8 Hz) are associated with drowsiness and inattention, while beta waves (13–30 Hz) correlate with active thinking and focus. Individuals with ADHD often exhibit an elevated TBR, where theta waves outnumber beta waves, reflecting a reduced capacity for sustained attention.
Clarke et al., 2011
The primary goal of this study was to design and evaluate a protocol to detect ADHD-related patterns using EEG data collected with Naxon Labs' Explorer tool and Muse headbands. The protocol included:
Baseline and Task Recording: Participants completed a baseline resting phase followed by a Go/No-Go task, a cognitive test commonly used to assess attention and impulse control. The Go/No-Go task is often used in psychology to measure an individual's cognitive abilities such as impulse control, attention, and reaction time. In this task the participant must refrain from performing an action when a visual stimulus tells them not to do, for example avoid pressing the space bar when the screen is red. Whilst wearing the Naxon Muse EEG headband, the participant completes a 5-minute period of baseline resting activity before moving on to complete a series of Go/No-Go trials. We
analyse the wave frequency in Naxon Explorer to see whether there appears to be a drastic difference in ratio between the theta and beta activity at baseline and during the trial. The EEG data is filtered to identify artefacts such as blinks and clenches which may disrupt the electrical activity recorded.
Data Analysis:
Thresholds for Interpretation:
Predictions -
- If there are symptoms of ADHD, the brain activity should be dominated by theta waves. This should be more pronounced in the Go/No-Go task than in the baseline.
- If the brain activity is typical, the beta waves should be more pronounced to show the brain is engaging in attentional control.
Findings -
- For regions AF7 and AF8 it appears the activity is not dominated by theta waves, more so by beta waves. Whereas for regions TP9 and TP10, the theta activity is much greater than beta activity.
Integrating artificial intelligence (AI) with EEG analysis offers a pathway to address these challenges. AI can detect subtle trends in noisy data, adapt models as new data becomes available, and provide personalized insights for both diagnosis and treatment. This approach could revolutionize ADHD management by offering tailored interventions such as neurofeedback.
There is evidently complications with the traditional approach to diagnosing ADHD which EEG can counteract. However, more still can be done to ensure diagnosis is robust and accurate across universal populations…
Using AI, biomarkers of ADHD can be detected from noisy EEG data and transformed into meaningful trends and patterns often missed by the human eye.
As more EEG data is collected from patients, AI models can be continuously refined. Machine learning techniques can be used to update the models as new data comes in, ensuring that the diagnostic tool remains up-to-date with the latest understanding of ADHD.
Clinicians and researchers can provide feedback on the AI’s performance, leading to iterative improvements in the model’s accuracy and reliability.
However extremely large sample sizes are needed to establish a tool which is not confounded by demographic characteristics such as co-morbid conditions.
Diagnosis is not the only focus! Combining AI with EEG can lead to the development of personalised treatment plans for patients with ADHD. Based on their individual patterns of brain activity, interventions such as neurofeedback can be tailored to help patients manage their symptoms.
As technology advances, integrating tools like EEG and AI into ADHD diagnosis and treatment raises important ethical questions. One concern is the increasing reliance on technology in clinical practice. Some clinicians argue that while these tools provide valuable insights, they should not replace the human element of diagnosis and treatment, which includes understanding the patient's lived experience and context.
Looking ahead, the potential for EEG and AI to autonomously diagnose and implement treatment plans sparks debate. While such innovations could streamline healthcare delivery, they also raise questions about the role of clinicians and the ethical implications of relying on machines to make critical health decisions.
Another contentious issue is the ability to "read minds" through brain imaging techniques like EEG. This raises concerns about privacy and consent, as participants may be wary of how their neural data could be interpreted or used beyond the intended scope of diagnosis.
Lastly, the uneven accessibility to advanced diagnostic tools poses significant challenges. Socioeconomic disparities may limit access to EEG and AI-based solutions, leading to unequal opportunities for accurate diagnosis and effective treatment. Addressing these ethical concerns is crucial to ensure that technological advancements in ADHD care are implemented responsibly and equitably.
.
This project underscores the potential of EEG technology, particularly the theta/beta ratio, as a tool for understanding ADHD. While there are limitations, the integration of advanced analytics, such as AI, could enhance the reliability and applicability of these findings. The work conducted during this project represents an important foundation for further research and development in ADHD diagnostics and personalized treatment.
For researchers, clinicians, and technologists, this collaboration between Naxon Labs and the University of Birmingham highlights the power of interdisciplinary innovation in addressing complex neurological conditions.
Abbie Bull is a Psychology student at the University of Birmingham. During her internship with Naxon Labs, she explored the potential of EEG-based diagnostics for ADHD, contributing to groundbreaking research in neurotechnology.
Naxon has financial support by ANII (National Research and Innovation Agency)
