The average clinical trial screen failure rate is 60%. That means for every 10 patients referred to your site, 6 fail screening β after coordinators have already invested time in contact, consent, and medical record collection.
That's your biggest hidden cost in patient recruitment.
AI Verdict: Likely Eligible β schedule pre-screening call
Turnaround: 24 hours from record receipt
What Hekma's AI Does
Aggregated via FHIR β every provider the patient has seen, unified into one record
Every inclusion and exclusion criterion from your trial protocol, automatically applied
Likely Eligible / Borderline / Likely Ineligible β with evidence for each determination
Flags patients that need one specific data point to confirm vs. patients that clearly do not qualify
Not the full record. Not a pile of PDFs. A structured verdict with the evidence behind it.
AI summaries generated within 24 hours of record receipt. Trained specifically on clinical trial I/E logic.
Before / After
| Without Hekma | With Hekma | Outcome |
|---|---|---|
| Manual record review β 3β6 weeks per cohort | AI summarisation β 24-hour turnaround per batch | Cut pre-screening time by 95% |
| 60% screen failure β coordinators wasted on ineligible patients | Only likely-eligible patients reach your sites | Screen failure: 60% β under 20% |
| Pile of PDFs from 5 different providers | One structured page per patient with criterion-level evidence | First call is a confirmation, not an investigation |
Compliance & Standards
AI summarisation pre-screens every candidate so your coordinators only talk to likely-eligible patients.