Independent clinical validation

We Find What Published Medicine Got Wrong.

The RocSite Discovery Engine identified that published ICU mortality estimates underestimate risk for elderly patients by 66–168% across three conditions simultaneously. Pre-registered. Cross-validated. 201,905 patients. Two independent datasets.

0
Patient encounters analyzed
0
Auto-certified findings
0
Institutions cross-validated
Engine running live Loading latest discovery…
What we do

Three ways to work with the Discovery Engine.

Service 01

Discovery Engine License

Dedicated engine instance against your proprietary data. Monthly contradiction reports. OSF pre-registration and evidence hashes for every finding. Your data stays in your environment.

$500K – $2M / yr Details →
Service 02

Adversarial Validation

Independent falsification testing of clinical AI models before FDA submission. Leakage detection, label stability, cross-dataset validation. Published methodology. Citable report.

$200K – $1M / engagement Details →
Service 03

Clinical Discovery Partnership

We run the engine against your cohort. Joint analysis, co-authored publication, OSF pre-registration before any analysis runs. All code published openly.

$100K – $500K + co-authorship Details →
Built on, cited by, and integrated with
medRxiv OSF PhysioNet FDA 510(k) Pre-Submission MIMIC-IV eICU-CRD
Featured finding

Myocardial infarction, elderly, ICU.
Published estimate off by 168%.

Across MIMIC-IV and eICU-CRD — two independent datasets — observed ICU mortality for elderly MI patients is 21.5%. The commonly cited published estimate is 8.0%. That's not a rounding error. It's a clinical risk stratification failure that shapes bedside decisions and every model trained on the literature.

Read the full finding →
Condition × AgeMI × >70
Observed (MIMIC-IV)21.5%
Published estimate8.0%
Divergence+168.4%
eICU confirmationn=12,408
Status● CONFIRMED
Evidence hash0x7a4e…c21b