Solutions / Cohort discovery

Medical imaging cohort discovery and feasibility

MedCorpora helps medical AI and life-sciences teams discover whether participating hospital archives may match a buyer-defined imaging cohort. It converts the requirement into aggregate inventory criteria, records verified and unresolved fields, and produces a feasibility decision before patient-level data moves.

Published 30 July 2026 · Reviewed 12 August 2026 · MedCorpora
InputBuyer cohort specification
First queryAggregate source inventory
OutputQualified feasibility decision
Next gateProgramme authorization
01

What the programme covers

Cohort discovery is the controlled step between a broad data request and a buildable programme. Search criteria can include modality, anatomy, pathology, acquisition, care setting, demographics, report evidence, pathology, outcomes and longitudinal history, subject to what each source can reliably expose at aggregate level.

02

What a useful specification includes

A defensible request defines the clinical task, source evidence and acceptance criteria before patient-level data moves. The exact fields and thresholds depend on the intended model claim.

  • Clinical question and intended model or research use
  • Inclusion, exclusion, positive and control definitions
  • Imaging protocol, scanner and quality constraints
  • Reference-standard and annotation requirements
  • Required clinical linkages and observation windows
  • Geography, site diversity, volume and timeline
03

Quality and validation controls

Discovery results retain uncertainty instead of turning approximate archive signals into promised cases. Definitions, source fields, counting units and known missingness are documented so buyers can judge whether a programme should proceed.

  • Patient, examination and series counting rules
  • Duplicate and repeat-examination treatment
  • Clinical-code and report-query precision
  • Required-field completeness
  • Verified, under-verification and sourcing-scope labels
04

Availability, rights and delivery

The deliverable is a scoped feasibility record and proposed programme design, not unrestricted access to hospital records. Approved curation and delivery are separate governed stages.

Public pages describe a sourcing and engineering capability, not guaranteed ready inventory. Each release remains subject to verified programme inventory, programme-specific authorization, privacy review, technical acceptance and buyer licence terms.

05

Questions, answered directly.

Can cohort discovery happen before data transfer?

Yes. Aggregate inventory assessment is intended to answer feasibility questions before patient-level data preparation.

Are discovery counts guaranteed final cohort counts?

No. Eligibility can change after permission, privacy, image-quality, linkage and ground-truth checks.

Can discovery include difficult negatives?

Yes when the buyer defines them and the source evidence can distinguish them reliably.

Can a buyer revise the specification?

Yes. Trade-offs can be evaluated before the programme is authorized and locked for execution.

Institutional engagement

Define the cohort.