Upload a patient's records and we check them against every recruiting trial on ClinicalTrials.gov. Same engine as the patient side, with no account attached.
No account, deliberately. Nothing is tied to your email or the patient's. You get a recovery code instead — it is the only way back to this search, so save it before you start.
De-identified before any model sees it. Names, dates of birth, record numbers and addresses are stripped on our server. The documents are never saved, and what we keep has no field that can hold a name.
Exclusion-driven. A trial is ruled out only on an explicit exclusion match. Inclusion criteria the record contradicts are surfaced on the trial rather than used to filter, because basket trials contradict most of their own cohort criteria by design.
Every criterion, separately. Each eligibility line is assessed on its own and recorded as met, not met or unknown, with the evidence used. Unknown means the record did not say — not that the patient fails it.
Not a clinical decision. This narrows the registry to a list worth your time. Eligibility is settled by the trial team at screening.
Anything counts. Clinic letters, pathology, genomic reports, discharge summaries. They need not be complete or tidy.
No blood results is fine. The search still runs, and we tell you exactly what was missing.
Identifying details are removed here. Names, dates of birth, record numbers and addresses come out on our server, before anything reaches a model.
Your documents are never saved. We keep a summary of the diagnosis and treatment, with no field that can hold a name.