BrainScope Ahead 100

Authorization

FDA created a new Class II EEG-based adjunct for recent mild traumatic brain injury assessment after a benefit-risk review.

Submission lineageDEN140025; no predicate submission.

Intended use

Provide adjunctive EEG-derived information when evaluating whether an adult with recent mild traumatic brain injury should receive head CT; it is not a CT substitute.

Input

Frontal raw EEG recorded within 24 hours of closed head injury; Computed quantitative EEG features

Output

Negative classification corresponding to activity consistent with no CT-visible structural injury; Positive classification that may occur with or without CT-visible structural injury; Raw qEEG measures for qualified interpretation

Use boundary

Adjunct only: a negative result does not replace CT when indicated, and a positive result does not establish a structural lesion.

AI disclosure and function

Algorithm / classifier

The De Novo summary describes a pre-established algorithm that filters EEG, reduces artifacts, computes and normalizes qEEG features and integrates them into a binary classification; it does not name an AI/ML architecture.

FDA AI-list inclusion and an EEG classifier do not establish an LLM, foundation model or generative-AI system.

No foundation-model, LLM, or generative-AI function was identified in the reviewed FDA materials.

Validation

Prospective, controlled, non-randomized, blinded, 11-site US pivotal study with CT and adjudicated clinical reference standards.

Comparator

Structural injury on head CT adjudicated by three blinded neuroradiologists, with prespecified clinical criteria for selected non-imaged patients.

Ground truth

Majority classification by a three-neuroradiologist panel for performed CT; protocol-defined New Orleans Criteria handling for selected GCS-15 patients without CT.

Representation

817 subjects were enrolled; 552 intended-use subjects were analyzed. The labeled population is age 18-80, but the reviewed summary section does not provide a complete demographic table. Numerous neurological and medical exclusions limit generalizability.

Reported performance

Sensitivity

78.5% (91/116; 95% CI 69.9%-85.5%)Prospective classification performance for CT-visible structural injury; the prespecified primary sensitivity endpoint was not met.

Specificity

48.6% (212/436; 95% CI 43.8%-53.4%)Prospective classification performance; the prespecified primary specificity endpoint was not met.

Negative predictive value

89.5% (212/237; 95% CI 84.8%-93.1%)Observed at 21% CT-positive prevalence in the analysis cohort; not transportable without regard to prevalence.

Positive predictive value

28.9% (91/315; 95% CI 23.9%-34.2%)A positive classification does not establish structural injury.

Limitations

• The pivotal study failed both predefined primary performance endpoints.
• 123 subjects had missing clinical or device classifications and were excluded from the diagnostic-accuracy analysis.
• The NPV depends on the 21% structural-injury prevalence in the study cohort.
• Safety and effectiveness were not evaluated in many clinically important groups, including prior brain surgery, brain tumor, seizure disorder and GCS below 13.

It does not replace CT, confirm injury after a positive result, meet the original pivotal primary endpoints or demonstrate improved patient outcomes.

PCCP / Change control

No authorized PCCP was identified for this submission in the reviewed FDA database record or decision material.

Absence of an identified PCCP does not prevent every future device change; it means no authorized plan was established in the reviewed public record.

Postmarket check

No exact-name openFDA device-enforcement result was found in the targeted query.

No exact-name openFDA result found as of 2026-09-08; this is not proof that no recall exists.

Primary FDA sources

1. FDA AI-enabled device list — Submission DEN140025Confirms inclusion, list display name, company, lead panel, product code and decision date only.
2. FDA submission database — De Novo DEN140025Confirms grant, review committee, product code and PCCP database field.
3. FDA decision summary — De Novo summary pp. 1-15; clinical study pp. 7-10; benefit-risk and special controls pp. 12-15Supports labeled population, algorithm description, prospective study, exclusions, failed endpoints, NPV interpretation and special controls.

Sources checked Sep 8, 2026 · Next scheduled review Oct 8, 2026. “No exact-name result” is a bounded search finding, not proof that no safety event or recall exists.