Back to search results

Improving Diagnostic Accuracy with AI Tools

This study is looking at how different ways of showing artificial intelligence (AI) predictions can help doctors and other healthcare professionals make more accurate diagnoses. It's testing two main approaches: one called "Bayesian-Updated Post-Test Probability," which combines a clinician's initial thoughts with AI information, and another called "Standard AI Predicted Probability," which just shows the AI's prediction. The study also looks at whether showing a "95% Confidence Interval" (a range that shows how sure the AI is) helps. The goal is to see if these methods improve diagnostic accuracy for patients with chest pain or shortness of breath. We are looking for 100 healthcare professionals, including Nurse Practitioners, Physician Assistants, and Doctors, who can complete an online survey in English.

Study design
This study is an interventional study with 100 participants. It uses a 2x2 factorial within-subjects design, meaning each participant will experience different ways of seeing AI predictions.
What's involved
Participants will complete an online survey using a computer or tablet. The primary endpoint, clinician diagnostic accuracy, is measured on Day 1 during survey completion.
Compensation
Not stated in the trial record.
Follow-up
Participants are followed for diagnostic accuracy, which is measured on Day 1 during survey completion.

AI-generated from the public study record. Only the study team can confirm whether you're eligible — confirm details with them before making decisions.

NCT07457840

Integrating AI Predictions With Clinician Expertise

Enrolling by Invitation
NAAges 18+Interventional
University of California, San Francisco
~100 participants
Updated 2026-07-14 on ClinicalTrials.gov
What's tested:Bayesian-Updated Post-Test ProbabilityStandard AI Predicted ProbabilityUncertainty Quantification (95% Confidence Interval)

At a glance

Recruiting sites
0 of 1 listed site is recruiting right now
RecruitingSuspended, closed, or not yet open
What they're measuring
Clinician Diagnostic Accuracy
Measured over Day 1 during survey completion

Conditions

Diagnostic Decision Making

Explore related trials

Chest PainDiagnostic Decision MakingDyspnea

Where it's being run

1 sites across 1 states
California1

Study leadership

  • Romain Pirracchio, MD, PhD, MPH · PRINCIPAL_INVESTIGATOR · University of California, San Francisco

Who to contact

This trial hasn't published a contact. View it on ClinicalTrials.gov

Do you actually qualify for this trial?

Add a private profile and we'll compare every criterion below against your situation — and tell you which ones are met, uncertain, or excluding.

Check eligibility for this trial ~2 min · HIPAA-protected · delete anytime
Eligibility criteria

Inclusion

Must hold one of the following clinical roles: Nurse Practitioner (NP), Physician Assistant/Physician Associate (PA), Resident Physician, Physician Fellow, or Attending Physician
Able to complete the survey in English
Access to a computer or tablet (mobile phones are not recommended due to the visual nature of the survey)

Exclusion

Does not hold an eligible clinical role as defined above
Completes fewer than 2 of 8 clinical vignettes (less than 25% of the survey)
Has previously participated in this study
Unable to complete the survey in English

What this trial measures

  • Clinician Diagnostic AccuracyDay 1 during survey completion

    Proportion of correct diagnostic assessments across all vignettes and experimental conditions. For each vignette, participants rate 5 possible diagnoses on a 0-100% probability scale. The diagnosis assigned the highest probability is considered the participant's final diagnosis. Accuracy is determined by comparing the final diagnosis to the ground truth diagnosis established by expert panel consensus (minimum 4 of 5 board-certified physicians in agreement). Analyzed using a generalized linear mixed model (GLMM) with binary outcome (correct vs. incorrect), fixed effects for CLR, uncertainty quantification, misleading AI, and vignette, and a random intercept for participant.