29/01/2026
đź§ New ADHD research worth noticing.
A study published in 2025 explored a key limitation in how ADHD is currently diagnosed. Most pathways still rely heavily on questionnaires, interviews, and rating scales. While useful, these tools are subjective, time‑intensive, and can vary widely depending on context and reporter.
This research used EEG data combined with machine‑learning techniques to support ADHD identification.
• A Random Forest model achieved 92.36% diagnostic accuracy.
• Strong precision, recall, and overall classification performance.
• The model reliably distinguished ADHD from control participants using brain‑based data.
The Australian diagnostic environment is under strain. Long waitlists, limited specialist access, and increasing scrutiny around ADHD diagnoses place pressure on clinicians, families, and available interventions.
Objective neurophysiological data may help strengthen ADHD assessment, rather than relying on behaviour reports alone. When considered alongside the Arns, Gunkelman, Breteler, & Spronk (2008) research into EEG predictors of medication outcomes it could point to a much faster process with greater first response and a reduced need for reviews.
This is not about replacing clinical judgment, but about giving clinicians better information to work with, and most importantly improving access and outcomes for those that need it.