Scalable Clinical Oversight of Large Language Models Via Uncertainty Triangulation
NCT07414966 · Status: NOT_YET_RECRUITING · Phase: NA · Type: INTERVENTIONAL · Enrollment: 7
Last updated 2026-02-17
Summary
This prospective, multi-reader, randomized crossover trial evaluates SCOUT (Scalable Clinical Oversight via Uncertainty Triangulation), a model-agnostic meta-verification framework that selectively defers unreliable large language model (LLM) predictions to clinicians by triangulating three orthogonal uncertainty signals: model heterogeneity, stochastic inconsistency, and reasoning critique. The trial assesses whether SCOUT-assisted review can reduce physician review time compared with standard manual review of AI-generated diagnoses while maintaining non-inferior diagnostic accuracy in coronary heart disease (CHD) subtyping.
Conditions
- Coronary Heart Disease (CHD)
Interventions
- DIAGNOSTIC_TEST
-
SCOUT-Assisted Review Workflow
SCOUT-Assisted Review (Intervention Arm): Physicians review 56 cases processed through the SCOUT framework. For cases classified as low-uncertainty (D(x)=0), the AI prediction is auto-accepted without physician review. For high-uncertainty cases (D(x)=1), the physician reviews the case with access to the main model's chain-of-thought reasoning and the meta-verification audit results. The main model is DeepSeek-V3.1 with chain-of-thought prompting.
- DIAGNOSTIC_TEST
-
Standard Manual Review Workflow
Physicians perform a full manual review of 54 cases using raw medical records with access to the AI model's predictions and reasoning, but without SCOUT uncertainty stratification or selective deferral.
Sponsors & Collaborators
-
China National Center for Cardiovascular Diseases
lead OTHER_GOV
Study Design
- Allocation
- RANDOMIZED
- Purpose
- DIAGNOSTIC
- Masking
- NONE
- Model
- CROSSOVER
Eligibility
- Min Age
- 18 Years
- Sex
- ALL
- Healthy Volunteers
- No
Timeline & Regulatory
- Start
- 2026-02-19
- Primary Completion
- 2026-02-28
- Completion
- 2026-02-28
More Related Trials
-
Manual Versus AI-Assisted Clinical Trial Screening Using Large-Language Models
NCT06588452 ·Status: RECRUITING
-
Comparison of Different Feature Engineering Methods for Automated ICD Coding
NCT04849195 ·Status: UNKNOWN
-
Physician Response Evaluation With Contextual Insights vs. Standard Engines - Artificial Intelligence RAG vs LLM Clinical Decision Support
NCT07037940 ·Status: COMPLETED ·Phase: NA
-
"Smart Family Doctor" Assisted Comprehensive Management of Secondary Prevention Among Post Revascularization Patients
NCT07273513 ·Status: NOT_YET_RECRUITING ·Phase: NA
-
Automated ICD Coding of Primary Diagnosis Based on Machine Learning
NCT04817423 ·Status: UNKNOWN
-
Physician Reasoning on Diagnostic Cases With Large Language Models
NCT06157944 ·Status: COMPLETED ·Phase: NA
-
The Application of Large Language Model in Emergency Chest Pain Triage
NCT06493175 ·Status: RECRUITING ·Phase: NA
-
The Diagnostic and Triage Capacity of Laypeople-large Language Model Collaboration in China
NCT07250516 ·Status: COMPLETED ·Phase: NA
-
Large Language Model for Understanding and Monitoring Elderly Neurocognition
NCT07347431 ·Status: RECRUITING
-
Research on Body Voice AI Recognition System for Children's Health Management
NCT06542120 ·Status: NOT_YET_RECRUITING
-
AI Models in Clinical Pathology Diagnosis: A Multicenter RCT
NCT07408167 ·Status: NOT_YET_RECRUITING ·Phase: NA
-
LLM-CoManage: Large Language Model-Enabled Co-Management of Hypertension, Diabetes, and Dyslipidemia
NCT07350486 ·Status: NOT_YET_RECRUITING ·Phase: NA
-
Atrial Fibrillation Recurrence Prediction
NCT06977516 ·Status: COMPLETED
-
Evaluation of AI Large Models for Diagnosis and Treatment in Real-World Cases: Multicenter Retrospective Study
NCT07378358 ·Status: RECRUITING
-
AI-Assisted Pathologist Performance Improvement: A Multicenter, Prospective, Randomized Controlled Trial
NCT07291362 ·Status: ENROLLING_BY_INVITATION ·Phase: NA
-
Artificial Intelligence for Learning Point-of-Care Ultrasound
NCT05900440 ·Status: ENROLLING_BY_INVITATION ·Phase: NA
-
Real-world Effectiveness Evaluation of Clinical Decision Support System Based on Artificial Intelligence (AI-CDSS)
NCT05065931 ·Status: COMPLETED
-
Artificial Intelligence - SARS-CoV-2 (COVID-19) Risk Evaluation
NCT04834934 ·Status: COMPLETED
-
AI-Assisted Interpretation of Cardiac CT in the Emergency Department
NCT07235657 ·Status: NOT_YET_RECRUITING ·Phase: NA
-
Artificial Intelligence-Enhanced Management for Coronary Heart Disease (AIM-CHD) : Impact on Cholesterol and Other CHD Risk Factors
NCT06686056 ·Status: COMPLETED ·Phase: NA
-
AI-Assisted Detection and Staging of Gastric Cancer Using Contrast-Enhanced CT
NCT07250347 ·Status: RECRUITING
-
Early Diagnosis of Clinical Evolution From SARS-CoV-2
NCT05787405 ·Status: COMPLETED
-
Exploring the Application Efficacy of Artificial Intelligence (AI) Diagnostic Tools in Medical Imaging (MI) of Respiratory(R) Infectious (I) Disease (D)
NCT06553911 ·Status: RECRUITING ·Phase: NA
-
Evaluating the Real World Performance of an AI Based Lung Nodule Detection Tool
NCT06597968 ·Status: RECRUITING
-
Feasibility of AI-based Classification of Normal, Wheeze and Crackle Sounds From Stethoscope in Clinical Settings
NCT05268263 ·Status: COMPLETED ·Phase: NA