Product direction
Institutional exam bank for research and AIThe exam archive, ready for research and AI.
Years of exams in legacy files and each vendor's formats become a governed institutional base: standardized, longitudinal, reviewed by specialists and with the source preserved.
Try it
From import to cohort.
ECG archive import
Legacy ECG system · 2009 to 2024
| Files read | 1,284,330 | XML, PDF and DICOM Waveform |
|---|---|---|
| Standardized in the canonical model | 1,271,904 | With signal, report and source |
| Duplicates reconciled | 8,112 | The same exam in two files |
| Report versions preserved | 3,457 | Amendments kept with their date |
| Awaiting review | 857 | Patient without a reliable identifier |
No record is silently discarded. Whatever does not reconcile stays visible for review.
Blinded review
Case C-0457 · ECG from 2021
Specialist A
Atrial fibrillation with rapid ventricular response
Specialist B
Atrial fibrillation with rapid ventricular response
92%agreement in the batch of 400 cases
Each reading is attributed to the specialist and kept separate from the original report.
Choose a question to see Kortex at work.
From source to use
How the archive becomes a base.
Legacy import
History enters through the same path as live systems, with the same checks.
Canonical model
Each exam arrives with its signal or document, its report and its source.
Attributed enrichment
Automatically extracted findings and specialist readings stay separate from the original text.
Governed use
Cohorts, controlled exports and access through APIs and MCP, by profile and purpose.
Start with the archive the institution already has.
We talk about the sources, the formats and the research questions to design the institutional bank.