Skip to content

Centific

Clinical trials do not fail only because of science.

Code Video Cerebral Valley

Core trick

Builds a per-trial rubric, scores physicians against five public sources, and links every verdict to the record it rests on.

Tags

healthscience

Description

Clinical trials do not fail only because of science. They fail when the wrong sites take the wrong trials, with weak patient fit, poor economics, and high patient burden. WEI helps practices quickly decide whether a trial fits their patients, workflow, and business. As trials get more complex and recruitment costs rise, this is increasingly critical for CROs, sponsors, and physician practices.
AkshatMittu/clinical-trial-matching★ 00 forksPythonCreated Jul 18, 2026 · last push Jul 20, 2026

README excerpt

# Clinical Trial Matching (Physician x Clinical Trial)

Matching physicians to clinical trials as potential **site investigators**, using
only public data, with every claim traceable to a source record.

Given a physician and a trial, it builds a rubric specific to that trial, scores
the physician against it from five public sources, asks the physician to close
whatever public data couldn't settle, and produces a report where every verdict
links to the record it rests on.

```
                    ┌── rubric ──┐
   trial record ────┤            ├──► adjudicate ──► interview ──► re-score ──► report
                    │  (per trial)         ▲          (asks)                      │
   physician ───────┴── evidence ──────────┘                                      │
   (5 public sources)                                                             │
                            comparable trials ──► precedent + patient proxy ──────┘
```

---

## Quick start

```bash
pip install -r requirements.txt
cp .env.example .env            # add your ANTHROPIC_API_KEY

# collect (no key needed, ~5 min)
python scripts/collect_trials.py          # 187 trials from ClinicalTrials.gov
python scripts/collect_physicians.py      # resolve investigators to NPIs
python scripts/collect_evidence.py        # 5-source dossiers + protocol PDFs

# score (needs key)
python scripts/build_rubrics.py --budget 2.00        # ~$0.86
python scripts/run_match.py --prebuild --budget 3.00 # ~$1.50

# the app
python scri

Team