AURORA-CRANIO is the deepest module in the pilot. Eight subsystems — origin biology, skull mechanics, biomarkers, diagnostics, surgical forge, neurodevelopment, benchmark, endpoints. It is more elaborate than any other module today, and it is the one we recommend prospective pilot teams look at first.
Why craniosynostosis
Three reasons. First, the disease is mechanically interesting: skull and brain growth are coupled, and that coupling is tractable to model. Second, the diagnostic frontier is moving toward photo-first and 3D-scan-first triage, which reduces ionising exposure in children — a real, measurable equity win. Third, the surgical landscape is rich: spring-assisted, endoscopic, open remodelling, and now generative planning all coexist.
What we got right
The biggest thing we got right was insisting on neurodevelopmental endpoints from day one. It would have been easier to ship a morphometry tool and call it done. The decision to commit to school-age neurodevelopmental outcomes as the actual endpoint pulled the rest of the module into shape.
What we got wrong
Two things. We underestimated the heterogeneity of single-suture vs syndromic cases, and our early benchmark suite over-indexed on isolated sagittal. We also shipped a forge UI that surgeons found too generative for the actual clinical conversation — it has been replaced with a more parsimonious comparison surface.