Spina bifida is the most common congenital defect of the central nervous system. AURORA-SPINA spans embryonic origin to adult life, with explicit support for prenatal surgical repair and continence/mobility outcomes.
A non-functional preview of the AURORA-SPINA surface. Composable panels, logged overrides, explainable end-to-end. Pilot sites tune this to their own workflow.
SPINA-ORIGIN Origin & Embryology72%SB-PRENATAL Prenatal Window91%SB-REPAIR Surgical Repair66%SB-PROTECT Neuro-Protection58%AURORA-SPINA is an open stack for the most common neural-tube defect compatible with life. Six subsystems span origin biology, the prenatal-window decision, surgical closure, lifelong protection, life-trajectory modelling and a discovery lane for new hypotheses.
Spina bifida is the most common neural-tube defect compatible with life. Outcomes pivot on timing — the prenatal window for closure, the postnatal CSF-diversion question, and the long arc of continence, ambulation and cognition. Fragmented systems track each chapter separately; AURORA-SPINA puts that whole timeline on one substrate.
The clinical evidence base is unusually rich. The MoMS randomised trial established prenatal repair as the standard of care for eligible patients, and its 30-month follow-up clarified the shunt-rate and motor benefits. Fetoscopic alternatives continue to mature. The questions in front of teams today are not whether to operate, but who, when and how — and how to support patients through decades of care.
AURORA-SPINA does not replace clinical judgement; it removes the substrate cost. Imaging, registry data and outcomes flow through one federated runtime. Decisions are explainable, overrides are logged, and patient data never leaves the institution.
We are not building a product around Spina Bifida. We are seeding an infrastructure — so that any hospital with the will to use it can.
AURORA-SPINA treats neural-tube closure as a single continuous trajectory with four intervention windows. Each window has its own subsystem; they share the same patient identity, the same audit log, the same consent metadata.
Every AURORA-SPINA 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.
Neural tube closure simulator with folate-axis perturbations.
Neural-tube closure simulator with folate-axis perturbations. Foundational priors that other SPINA subsystems consume.
Imaging, lesion-level and timing decisions for in-utero repair.
Imaging and lesion-level analysis for the prenatal window. Produces a structured decision packet for the fetal-medicine team.
MOMS-aligned planning for open and fetoscopic closures.
MOMS-aligned planning for open and fetoscopic closures, including approach selection and risk-adjusted timing.
Long-term shunt, cord and skin protection over the lifespan.
Lifelong shunt, cord and skin protection — a longitudinal companion that does not let the case fall off the radar.
Continence, ambulation, cognition and quality-of-life modelling.
Continence, ambulation, cognition and quality-of-life modelling, with explicit uncertainty bands.
Open hypotheses, registry signals and replication tracker.
Open hypothesis tracker, registry signals and replication watch — where new SPINA research is announced first.
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-SPINA.
Neural-tube closure simulator. Encodes folate-axis perturbations, hyperthermia, anticonvulsant exposure (valproate, carbamazepine) and genetic associations (VANGL1/2, PAX3, MTHFR variants). Not patient-facing on its own; supplies priors that condition every other SPINA subsystem.
Imaging + biochemical synthesis for the prenatal window. Reads targeted ultrasound, fetal MRI and AFP / AChE; outputs lesion type and upper boundary, ventricular atrium width, hindbrain-herniation grade, and a MoMS eligibility check. The output is a packet, not a recommendation — the team and the family decide.
Surgical-repair planner. MoMS-aligned for open hysterotomy; current-evidence-aligned for fetoscopic (single-port and two-port). Produces approach-specific risk envelopes — premature rupture of membranes, preterm birth, dehiscence, oligohydramnios — and explicitly flags where the evidence is thin or single-site.
Lifelong protection module. Shunt-failure forecast (12 / 30 / 60 month), tethered-cord recurrence watch, skin-breakdown risk from pressure-mapping data, and latex-allergy exposure log. Designed to be the case's longitudinal home — the place where the patient does not fall through the cracks between paediatric and adult services.
Trajectory model. Joint multi-output regression over ambulation (Hoffer scale 1–5), continence (urodynamic phenotype + caregiver-reported), cognition (PIQ / VIQ, BRIEF executive function) and quality of life (PedsQL spina bifida module). Uncertainty bands widen honestly when input is sparse.
Open hypothesis tracker. Surfaces emerging registry signals, replication attempts and disagreements between cohorts. RFCs against any SB-* subsystem must cite SB-DISCOVER entries before changing endpoints, thresholds or eligibility logic.
6 drafts by the module team and pilot collaborators. Each documents one slice of the substrate — methods, evaluation protocol, lessons. Slot PDFs in as they're ready; the entries point to placeholders until then.
These are the endpoints AURORA-SPINA is built to measure across pilot deployments. The targets below are pilot goals, not retrospective results.
Across registered prenatal-repair cohorts (target lift).
Reduction in CSF diversion under aligned in-utero repair.
Auto-staged vs. expert fetal-medicine consensus.
Composite urodynamic + caregiver-reported score.
Every override logged with reason.
From scan to family-ready packet.
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.
# 1 · install pip install aurora-spina # just AURORA-SPINA # or pip install aurora-neuro[peds] # paediatric bundle (SPINA included)
# 2 · verify aurora doctor spina # → AURORA-SPINA ✓ python ✓ models ✓ datasets # → six subsystems loaded # → ready in target ≤120ms
# 3 · try a public case
from aurora.spina import load_case
case = load_case("bench/spina/case-0021")
print(case.lesion_level) # → "L4"
print(case.repair_window) # → "prenatal (24+1)"
print(case.urodynamic_band) # → "moderate · σ 0.18"AURORA-SPINA 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.
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.
Most pilot sites land on AURORA-SPINA 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.
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-SPINA is independently downloadable there.
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.