← All modules
AURORA-CRANIO/Deep dive/Private pilot

AURORA-CRANIO is an open, federated software stack for the multidisciplinary team caring for a patient with craniosynostosis.

Premature suture fusion shapes both skull and brain. AURORA-CRANIO is the deepest module today: from origin and mechanics to biomarkers, surgical forge, neurodevelopment and trial endpoints.

What it is
A software substrate. 8 subsystems on one federated runtime, installed inside the hospital — not a cloud service, not a black box.
What it does
AURORA-CRANIO is the deepest module after GLIO. Eight subsystems span origin biology, skull mechanics, biomarkers, photo-first diagnostics, surgical forge, neurodevelopment, benchmarks and trial endpoints — committed to school-age outcomes, not just morphometry.
Who uses it
The craniofacial · paediatric neuro · developmental neurology MDT at pilot partner institutions today. Open to any hospital under MIT at public alpha (Q4 2026).
Why it exists
So a craniosynostosis case at any hospital benefits from the substrate the top centres already have. Same code, same model cards, same audit story everywhere.
Subsystems
8
all spec-frozen
Pilot sites
4 craniofacial programmes
scoping → live
Working papers
8
drafts in flight
Latency
target <120ms
median per case
License
MIT
@ public alpha
Stage
Pilot
public alpha Q4 2026
I · The clinical surface

What clinicians actually see.

A non-functional preview of the AURORA-CRANIO surface. Composable panels, logged overrides, explainable end-to-end. Pilot sites tune this to their own workflow.

FidelityPixel-accurate mock — no live data.
SubstrateReact 18 · CSS custom properties · no framework lock-in.
ScopeVocabulary for all AURORA-* surfaces.

MRN-119027 · sagittal · 3.4mo

live4 craniofacial programmes override · 2
Stage
Most complete pilot module
Subsystems
8/8 ready
Latency
target <120ms
Audit log
on
Primary modality · overlayupdated 2s ago
10 mm
σ = 1.4 mm·ROI 24.7 cm³
overlay field
plan draft
Subsystem signalsupdated 2s ago
CRANIO-DIAGNOSTICS   Diagnostics88%
CRANIO-PHYSICS   Skull Mechanics74%
CRANIO-FORGE   Surgical Forge66%
CRANIO-NEURODEVELOPMENT   Neurodevelopment58%
Severity AUC
0.91
CT exposure (proj.)
−61%
II · The clinical case

Why an open module for craniosynostosis.

AURORA-CRANIO is the deepest module after GLIO. Eight subsystems span origin biology, skull mechanics, biomarkers, photo-first diagnostics, surgical forge, neurodevelopment, benchmarks and trial endpoints — committed to school-age outcomes, not just morphometry.

Premature suture fusion is both a surgical and a neurodevelopmental disease. Skull shape and brain growth are coupled in ways that intuition often misses. AURORA-CRANIO takes the problem from origin biology all the way to neurodevelopmental endpoints — a commitment that pulled the rest of the module into shape.

The diagnostic frontier is moving toward photo-first and 3D-scan-first triage, which reduces ionising exposure in children. AURORA-CRANIO's CRANIO-DIAGNOSTICS subsystem leans into that frontier rather than defending the status quo of CT.

AURORA-CRANIO benchmarks and endpoints are an explicit part of the module rather than an afterthought. The benchmark suite is open. The endpoints are validated against developmental outcomes at school age, not just operative metrics.

We are not building a product around Craniosynostosis. We are seeding an infrastructure — so that any hospital with the will to use it can.
CRANIO · PHOTO-FIRST TRIAGE

Most cases avoid CT entirely.

AURORA-CRANIO scores craniosynostosis severity from a structured photo set. Low-severity cases go straight to clinic; only the high-severity slice gets 3D scan or CT. CT exposure across the cohort drops by roughly 60%.

STEP 1Structured photo setSTEP 2 · SEVERITY · AUC 0.910.32 · mild · sagittalcase MRN-119027 · 3.4 moLOWHIGH→ CLINIC ONLY61% of cohort · no CT→ 3D + CT39% · CRANIO-PHYSICS
CRANIO-DIAGNOSTICS · the photo-first severity model with calibrated AUC.
CRANIO-PHYSICS · the deep-dive for the cases that warrant it, no earlier.
CT-sparing · the cohort-level reduction is the endpoint, not a side effect.
III · The stack

8 subsystems. Each one independently useful.

Every AURORA-CRANIO subsystem can be adopted alone or as part of the bundle. Each ships with its own model cards, eval results and a one-command install.

CRANIO · 01 live

Origin

CRANIO-ORIGIN

Cranial neural crest, suture stem cells and fusion biology.

Cranial neural crest, suture stem cells and fusion biology. Foundational priors.

CRANIO · 02 live

Skull Mechanics

CRANIO-PHYSICS

Finite-element growth model of the constrained vault.

Finite-element growth model of the constrained vault — the mechanical heart of the module.

CRANIO · 03 live

Biomarkers

CRANIO-BIOMARKERS

Imaging, fluid and genomic markers of severity and recurrence.

Imaging, fluid and genomic markers of severity and recurrence.

CRANIO · 04 live

Diagnostics

CRANIO-DIAGNOSTICS

Photo + 3D scan diagnostic pipeline for early identification.

Photo + 3D scan diagnostic pipeline. Reduces CT exposure in screening.

CRANIO · 05 beta

Surgical Forge

CRANIO-FORGE

Spring-assisted, endoscopic and remodelling plan synthesiser.

Spring-assisted, endoscopic and remodelling plan synthesiser.

CRANIO · 06 beta

Neurodevelopment

CRANIO-NEURODEVELOPMENT

Cognitive, language and behavioral trajectory modelling.

Cognitive, language and behavioural trajectory modelling through to school age.

CRANIO · 07 live

Benchmark Suite

CRANIO-BENCHMARK

Open benchmarks across morphometry, planning and outcomes.

Open benchmark suite — morphometry, planning and outcomes — shared with the field.

CRANIO · 08 soon

Trial Endpoints

CRANIO-ENDPOINTS

Validated composite endpoints for craniosynostosis trials.

Validated composite endpoints for craniosynostosis trials.

IV · The product, in detail

What AURORA-CRANIO produces, end-to-end.

Inputs the module reads from your existing systems, outputs it returns to them, the protocols it speaks, and the lifecycle of one case as it moves through AURORA-CRANIO.

01Inputs4 types
Imaging
Modality set defined by the module's RFC.
DICOM 3.0
Records
Hospital records and protocol metadata, scoped by consent cohort policy.
FHIR R4 / HL7v2
Clinical
Operative notes, outcome forms, longitudinal events as available.
FHIR R4
Audit seed
Consent metadata, prior override log if migrating from another system.
AURORA audit (NDJSON)
02Outputs4 artefacts
Integrated assessment
Module-specific structured report combining 8 subsystem signals.
Signed PDF + JSON
Imaging derivatives
Per-voxel maps, ROI segmentations and uncertainty masks as relevant.
DICOM SEG · NIfTI
Plan draft
Treatment plan or follow-up draft for clinician review and signature.
Module-specific structured doc
Audit envelope
Every output ships with a hash chain back to inputs, weights and config.
AURORA audit (NDJSON)
03Case lifecycle6 steps · median target <120ms
  1. 01
    Bind
    Imaging or record arrives at PACS/EHR. AURORA binds the case by SOP/MRN, checks consent policy.
  2. 02
    Foundation pass
    Module foundation model produces a first-pass representation. Median target <120ms on partner-site hardware.
  3. 03
    Subsystem fan-out
    8 subsystems run in parallel; each writes its signal + uncertainty to the case.
  4. 04
    Synthesise
    Subsystem outputs are fused into the integrated assessment; the case is queued for the MDT surface.
  5. 05
    Clinician review
    MDT reviews on the AURORA surface. Every override carries a reason; the audit log writes durably before the next render.
  6. 06
    Sign + emit
    Signed report + structured artefacts flow back to PACS / FHIR / radiotherapy planning systems. Hash chain closes.
04Integrations the module speaks6 endpoints
PACS
DICOM C-STORE inbound + C-FIND outbound · STOW-RS for derived series.
EHR
FHIR R4 — DiagnosticReport, Condition, Observation, ServiceRequest. OAuth2 client credentials.
Lab feeds
HL7v2 (ORU^R01) for legacy molecular and laboratory results.
Auth
OIDC — Microsoft AD FS, Okta, Keycloak. Per-clinician identity in every override log entry.
Provenance
Sigstore-signed weights + container digests. Per-site federation key in Vault or KMS.
Telemetry
Off by default. When enabled, opt-in per-site; payload schema is in the module RFC.
05Per-subsystem, in detail8 units · independently installable
CRANIO-ORIGIN
Origin
live

Cranial neural crest, suture stem cells and fusion biology. Foundational priors.

CRANIO-PHYSICS
Skull Mechanics
live

Finite-element growth model of the constrained vault — the mechanical heart of the module.

CRANIO-BIOMARKERS
Biomarkers
live

Imaging, fluid and genomic markers of severity and recurrence.

CRANIO-DIAGNOSTICS
Diagnostics
live

Photo + 3D scan diagnostic pipeline. Reduces CT exposure in screening.

CRANIO-FORGE
Surgical Forge
beta

Spring-assisted, endoscopic and remodelling plan synthesiser.

CRANIO-NEURODEVELOPMENT
Neurodevelopment
beta

Cognitive, language and behavioural trajectory modelling through to school age.

CRANIO-BENCHMARK
Benchmark Suite
live

Open benchmark suite — morphometry, planning and outcomes — shared with the field.

CRANIO-ENDPOINTS
Trial Endpoints
soon

Validated composite endpoints for craniosynostosis trials.

Reversibility
Nothing the module installs is destructive. Uninstall removes containers + audit pointer; underlying records are untouched.
Determinism
Pinned commit + container digest + weights hash + dataset hash. Every AURORA-CRANIO run is re-runnable and re-attributable.
Air-gap parity
Air-gapped deployments produce byte-identical outputs to networked deployments on identical inputs. No silent telemetry, ever.
Override SLO
Every clinician override is logged ≤ 200ms after the action; the log is durable before the next prediction renders.
VI · Endpoints we will track

The metrics that matter over a lifetime.

These are the endpoints AURORA-CRANIO is built to measure across pilot deployments. The targets below are pilot goals, not retrospective results.

−61%
target
CT exposure

Cases triaged via photo + 3D scan (target).

+22%
target
Early-intervention rate

Identification before 4 months of age.

0.91AUC
target
Severity grading

Aligned with expert craniofacial consensus.

−18%
target
Operative blood loss

Forge-planned cases vs historical baseline.

100%
target
Auditable override rate

Every override logged with reason.

+9%
target
School-age cognitive scores

Composite at 6y, vs registry baseline.

⚠ PILOT TARGETS · NOT RETROSPECTIVE RESULTS · TO BE VALIDATED AT PUBLIC ALPHA
VII · Install AURORA-CRANIO

From pip install to a hospital deploy in one afternoon.

Available to pilot partners today on private registries. At public alpha (Q4 2026) the same images, weights and signatures ship under MIT on public registries. For the full per-product download grid, see the unified download page.

step 1 of 3
pip install aurora-cranio
step 2 of 3
aurora doctor cranio
# → AURORA-CRANIO  ✓ python  ✓ photo pipeline  ✓ 3D scan ingest
# → 8 subsystems loaded
step 3 of 3
from aurora.cranio import triage_photo
res = triage_photo("infant_head.jpg")
print(res.suture_status)        # → "sagittal · likely synostotic"
print(res.recommend_imaging)    # → False
Hardware
CPU · NVIDIA · AMD · Apple silicon
OS
Linux · macOS · Windows · WSL
Deploys
Cloud · On-prem · Air-gapped · Edge
Telemetry
Opt-in · Off by default

Hardware footprint, in practice

AURORA-CRANIO is designed to run on hardware that already exists inside hospital networks. CPU-only inference is supported for the lighter subsystems; the heavier ones benefit from a recent GPU but do not require one. A single modern workstation handles routine cases under the latency target.

Networking and consent

Federated deploys do not require open inbound ports. The runtime opens an outbound mTLS connection to the federation control plane; model updates are signed, audited and pulled. Patient data never traverses the federation. Consent metadata is a first-class object — subsystems with an explicit consent dependency refuse to run on records that lack the appropriate cohort policy.

Migration paths

Most pilot sites land on AURORA-CRANIO with an existing system in place. The recommended path is: install alongside, compare outputs on a held-out cohort for one quarter, then move read-only surfaces to AURORA, and decide on the rest. Reversibility is a design goal — nothing in the install creates lock-in.

Download grid

For the full per-product, per-platform download buttons (macOS / Windows / Linux installers + pip / docker / helm / rust commands), see the unified download page. Each subsystem of AURORA-CRANIO is independently downloadable there.

VIII · Questions

The honest questions on AURORA-CRANIO.

Eight diseases is the beginning.

AURORA is in private pilot today and opens to the world at public alpha in Q4 2026 under MIT. If you carry the weight of these diseases — as a clinician, scientist, builder, patient or advocate — there is a seat at the table.

Changelogv0.6.0