Neurora has received FDA approval for an AI-based test designed to identify indicators of autism spectrum disorder in children between the ages of three and five, the company announced this week. The system analyzes electroencephalogram data and is intended to support, not replace, clinical diagnosis.

Early identification matters because interventions tend to be more effective when they begin sooner, and current diagnostic pathways often involve long waits for specialist assessment. A tool that can flag children for further evaluation earlier in that process addresses a real bottleneck.

What the test measures

EEG-based approaches look for patterns in brain activity rather than relying solely on behavioral observation. That has an obvious appeal: behavior can vary by context, by who is observing and by a child’s comfort in a clinical setting. A physiological signal is, in principle, more consistent.

The caveat is that consistency in a signal does not guarantee diagnostic accuracy across populations. Performance can differ by age within the approved range, by co-occurring conditions and by the quality of the EEG recording.

Approval is not the end of validation

FDA approval establishes that the test meets a regulatory standard for safety and effectiveness in its stated use. It does not answer every question clinicians will have: how it performs in community settings versus research environments, how often it produces ambiguous results, and how its output should be weighed alongside behavioral assessments.

Post-market studies and independent evaluations will matter more to adoption than the approval announcement.

The equity question

Any screening tool that requires EEG equipment and trained staff faces access constraints. If the test is primarily available in well-resourced pediatric centers, it may speed diagnosis for families who already have relatively good access while leaving others waiting. The clinical value and the distribution problem need to be solved together.