Neurology · Quantitative brain MRI volumetry and neurodegeneration decision-support
NeuroQuant
Cortechs.ai (formerly CorTechs Labs, Inc.)
NeuroQuant is the longest-standing quantitative brain-MRI product on the US market — first FDA-cleared in the 2006-2007 review cycle, years before "AI medical device" was a regulatory category, and the named predicate that later volumetry engines such as icobrain were cleared against. Eighteen years on it carries a broad regulatory footprint (FDA, a European CE mark, and a Health Canada licence) and the deepest deployment record in its category, though the headline reach figures are company-reported. The independent evidence base is genuinely substantial and, unusually, includes a cautionary result: across the best-studied indication — hippocampal-volume asymmetry for mesial temporal sclerosis — the software matches or approaches expert neuroradiologists in some cohorts yet was less sensitive than expert visual reading in another. That evidence is retrospective rather than randomised, which is what holds NeuroQuant at two marks: a wide approval footprint, an exceptional deployment track record and a clean public safety record earn the second mark, and the absence of randomised evidence, together with approvals that are not all confirmable in a public register, keep it from the third.
Performance Metrics
Clinical Evidence
NeuroQuant's evidence base is strongest in epilepsy and dementia, and its most useful feature for a procurement reviewer is that the independent literature reports both favourable and unfavourable results — this is not a one-sided vendor dossier. In epilepsy, two independent studies frame the range. Azab et al. (AJNR 2015; PMID 25907519) compared NeuroQuant's automated hippocampal-asymmetry index against neuroradiologists' visual reads in 63 patients with proven mesial temporal sclerosis and found the software's standalone accuracy (79.4%) comparable with trained neuroradiologists (75.0-84.2% depending on the asymmetry threshold), while running unattended in five to ten minutes per study. The larger and more cautionary study is Louis et al. (AJNR 2020; PMID 32217554), which used surgical pathology as the reference standard in 144 patients evaluated for temporal-lobe epilepsy: here NeuroQuant was less sensitive than expert visual MRI reading (sensitivity 69.0% versus 93.0%) with a lower negative predictive value (79.8% versus 95.0%), though its positive predictive value was comparable (84.0% versus 89.1%). The honest reading is that NeuroQuant standardises and quantifies hippocampal asymmetry usefully but does not replace an expert read, and a negative quantitative result does not exclude sclerosis. In dementia and Alzheimer's disease the software's hippocampal and whole-brain volumetry is used as an objective adjunct to visual atrophy rating. Independent comparisons against medial-temporal-lobe visual scales report high correlation between the two approaches (for example the automated-versus-visual comparison in PMID 29172642), and a 2025 academic-memory-clinic evaluation (PMID 40309995) examined the clinical value of NeuroQuant hippocampal volumes and hippocampal- occupancy scores in routine practice; reported diagnostic sensitivity and specificity for objective memory impairment varied by threshold and cohort. For field context, meta-analyses of hippocampal volumetry in general (not NeuroQuant specifically) place pooled sensitivity and specificity for Alzheimer's disease in the low-to-mid 80s percent and materially lower for mild cognitive impairment — a reminder that volumetry supports, rather than settles, the diagnosis. The structural caveat across all indications is the same as for the rest of the quantitative-neuroimaging field: no randomised controlled trial of NeuroQuant has been published as of the review date, and the strongest validations are retrospective single- or multi-centre cohorts. The value proposition is reproducible, reader-independent measurement and longitudinal tracking, not autonomous diagnosis.
| Study | Design | n | Sensitivity | Specificity | AUC | Published |
|---|---|---|---|---|---|---|
| Azab M, Carone M, Ying SH, Yousem DM (independent; Johns Hopkins) | RetrospectiveRetrospective | 63 | — | — | Accuracy 79.4% (NeuroQuant) vs 75.0-84.2% (neuroradiologist, by asymmetry threshold) | AJNR Am J Neuroradiol, 2015; 36(8):1400-6; PMID 25907519 |
| Louis S, Morita-Sherman M, Jones S, et al. (independent; Cleveland Clinic; pathology reference standard) | RetrospectiveRetrospective | 144 | 69.0% (vs 93.0% expert visual read) | — | PPV 84.0% vs 89.1%; NPV 79.8% vs 95.0% | AJNR Am J Neuroradiol, 2020; 41(4):591-7; PMID 32217554 |
Clinical Pulse
Verified clinician sentiment
Verified clinician reviews launching soon · Apply to contribute →
Reviews are signed by clinicians verified against their regulator's public register.
Inside the Auris+ Listing
Five more sections complete this device’s Auris+ Listing.
Decision Ledger
ProPro unlocks a private, cross-vendor log of every case you read on this device — what the AI called, what you concluded, and a one-line reason if you overrode it. Ready for the EU AI Act's deployer logging obligations when they land in 2028.
Clinical Evidence Deep Dive
ProPro unlocks the structured clinical-evidence summary — study count, target patient population, and a tabular accuracy-metrics view drawn from peer-reviewed sources.
Peer-Reviewed Publications
ProPro unlocks the curated peer-reviewed publication list with PubMed cross-links — the citation backbone of every editorial verdict.
Post-Market & Regulatory Conditions
ProPro unlocks the post-market surveillance summary, recall record, and the conditions of approval that bound real-world use.
AI Algorithm Version History
ProPro unlocks the chronological record of algorithm version changes — what changed when, drawn from manufacturer changelogs and regulatory filings.
Regulatory Approvals
K241098
Class II
Safety Record
No Cortechs.ai / CorTechs Labs or NeuroQuant device recall or FDA safety communication was identified in publicly available sources as of July 2026, and a search for NeuroQuant adverse-event reports surfaced none. This should be read as "none found in public reporting" rather than an exhaustive audit: the FDA MAUDE and enforcement databases and the openFDA API could not be queried directly from this environment (egress policy blocks the FDA hosts), so the safety record rests on secondary reporting and the absence of any recall announcement. NeuroQuant is a measurement and reporting overlay — it quantifies structures and lesions and compares them with a normative reference but does not modify source images, remove studies from the reading worklist or drive treatment automatically — which bounds the harm surface. The principal documented risk is over-reliance: independent epilepsy validation found the software less sensitive than expert visual reading for hippocampal sclerosis, so a quantitative output, or the absence of a flagged abnormality, does not substitute for the radiologist's independent read.
Intended Use & Indications
NeuroQuant is a software-only medical device intended for automatic labelling, visualisation and volumetric quantification of segmentable brain structures from a set of MR images. In routine use the software ingests a 3D T1-weighted brain MRI (and, for lesion work, FLAIR and susceptibility/T2*GRE series), segments the structures of interest, and reports absolute and normalised volumes together with the patient's percentile position against an age- and sex-matched normative database — the manufacturer states applicability across ages 3 to 100 using its Dynamic Atlas reference model. Where a prior study is available the software aligns the two and reports the rate of volume change over time, the capability most relevant to tracking neurodegeneration and treatment response. Clinically the tool is applied to hippocampal and whole-brain atrophy in dementia and Alzheimer's disease, hippocampal-volume asymmetry in the work-up of mesial temporal sclerosis in epilepsy, lesion and atrophy assessment in multiple sclerosis, and, in the NeuroQuant 5.0 release, segmentation and quantification of microhaemorrhages and amyloid-related imaging abnormalities (ARIA-E and ARIA-H) in patients receiving amyloid-beta-directed antibody therapy. In every setting the device is decision support: it standardises and quantifies measurements that a clinician would otherwise estimate visually, and the interpreting physician retains final diagnostic responsibility.