Tuyen Tran, MD
Research

When does clinical AI actually work?

A clinical AI reports 90% accuracy in trials. Drop it into your hospital and that number shifts. I study when the evidence holds and when it breaks — across operators, settings, and benchmarks.

Pipeline 2026

16items in motion

Publicly released

31%5 / 16
Published1
Preprint2
Open source2
Under review5
In progress4
Patent2
Interactive demo

Same evidence. Two very different numbers.

An AI diagnostic tool reports 90% accuracy across 4,000 cases. That's the average — pooled across every operator. Slide to see how the conditional accuracy shifts.

Reported (pooled)90%
Conditional on operator88%

Gap: 2 percentage points lower than the reported figure.

Operator

Junior attending

skill index = 72

TraineeExpert

Numbers in this demo are illustrative — not actual study results.

Conditional validity — clinical evidence holds only conditional on who runs it, where, and against which benchmark. A 90% pooled accuracy can mean 50% in trainee hands and 96% in expert hands. Most validation studies don't report this. That's the programme.

"Most validation studies report one number. The truth has dozens."

Three directions

Direction A

Epistemic drift

AI increasingly produces the evidence we rely on. Measuring and naming what silently drifts.

Direction B

Operator-dependence + LMIC

Surgical and diagnostic trials don't transfer cleanly between settings. Measuring what actually transports.

Direction C

Deployment-validity index

A structured index for whether a piece of clinical AI evidence should be trusted in a given setting.

Work

16 items · 4 groups
01

Clinical AI & epistemology

When models help produce evidence — measurement and evaluation frameworks.

3 items
  • Structural Validity Without Direction Fidelity: Evaluating LLM-Generated Medical Abstracts Across Model Scales

    Target: NEJM AI

    Two-paper series, data collection complete (n = 1080 + 720 + 120). OSF prereg X4RP5.

    In progress
  • When Coherence Is No Longer Evidence: Epistemic Drift in AI-Augmented Science

    Accountability in Research · Q1

    Under review
  • Epistemic Immunodepression in Clinical Evidence

    MetaArXiv · CC-BY 4.0 · 10.31222/osf.io/gqunf_v1

    Cross-posted to LessWrong.

    Preprint2026
02

Method frameworks & patents

OPERA · CIVER · BRIM — original rubrics in progress.

4 items
  • OPERA — Evidence Framework for Conditional Validity in Operator-Dependent Medicine

    Target: Journal of Clinical Epidemiology · Q1

    Companion framework to the M2 pilot. v6 draft.

    In progress
  • A pilot of a new clinical evidence framework: conditional validity in operator-dependent research

    MetaArXiv · pending DOI

    OPERA empirical pilot. κ = 0.965 across 4 paired surgical case studies.

    Preprint2026
  • CIVER — Clinical Information Veracity Evaluation Rubric

    Vietnam patent · filed 2026-03-27

    Patent
  • BRIM — sister rubric framework

    Vietnam patent · filed

    Patent
03

Pediatric & plastic surgery

Anorectal and urogenital malformations, clinical case work.

7 items
  • Published2026
  • Staged vs Single-stage Anorectal Malformation Repair — a Meta-analysis

    Pediatric Surgery International · Q2

    Under review
  • Transperineal Ultrasound Transferability Between HIC and LMIC ARM Cohorts

    Target: Pediatric Radiology

    ARM cohort transferability seed paper.

    In progress
  • Frozen Section Biopsy in Hirschsprung Disease

    Journal of Indian Association of Pediatric Surgeons · Q3

    Under review
  • Gastric Volvulus in a Pediatric Patient — Case Report

    Journal of Surgical Case Reports

    Under review
  • Single-stage Repair of Proximal Hypospadias — Case Series

    Urology Case Reports · Q3

    Under review
  • Tiered Surgical Strategy for Posterior Urethral Valves in an LMIC Setting

    Target: Journal of Pediatric Urology · Q1

    Three-arm cohort (n = 27), 2015–2023.

    In progress
04

Open-source software

CiteCheck · AI for Academic — tools released free to the community.

2 items
  • CiteCheck

    Python · PyPI · MIT licence

    Reference verification engine. Powers the AI for Academic Paper Checker.

    Open source
  • AI for Academic

    Web · aiforacademic.world

    12-tool research toolkit. Partially open source.

    Open source

Want to collaborate or ask a research question?

Book an advisory session →