AI-Assisted Medical Decision-Making

A Cohort Study to Evaluate an Artificial Intelligence Model for Assisting Medical Decision-Making Using Real-Time Hospital-Wide Electronic Health Record Data

The study builds and applies an AI model to help doctors predict patient diagnoses and outcomes, such as survival or hospital stay. Real-time, multimodal data (labs, vital signs, history, imaging) from hospital records will be used. Patients will be tracked to compare the AI's performance with standard care. The goal is to improve diagnosis and treatment accuracy in a real-world, prospective study.

Study Overview

• Status: Recruiting
• Conditions: Real-world Study
• Intervention / Treatment: Other: AI-associated strategy

Detailed Description

This study aims to build and apply an artificial intelligence (AI) model to assist doctors in predicting patient diagnoses and outcomes, such as survival or hospital stay length. Patients will be enrolled across the hospital, and real-time, multimodal health data-including lab results, vital signs, medical history, and imaging-from electronic health records will be used. The study will follow participants to evaluate the AI model's performance against standard practice. The goal is to improve the accuracy and speed of diagnoses and treatments, enhancing patient care. This prospective study tests the model in real-world hospital settings.

• Study Type: Observational
• Enrollment (Estimated): 50000000

Contacts and Locations

• Study Contact: Name: Fei Liu, MD; Phone Number: +86 13810512704; Email: liufei_2359@163.com

Study Locations

• China
  • Zhejiang
    • Wenzhou, Zhejiang, China
      • Recruiting
      • First Affiliated Hospital of Wenzhou Medical University
      • Contact: Cheng Tang; Phone Number: +86-0577-88002888; Email: c249325687@163.com
    • Wenzhou, Zhejiang, China
      • Recruiting
      • Second Affiliated Hospital of Wenzhou Medical University
      • Contact: Linfeng Liu; Email: llfwfe@163.com

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Child; Adult; Older Adult
• Accepts Healthy Volunteers: Yes

• Sampling Method: Non-Probability Sample

Study Population

This study includes all patients admitted to any hospital department, with real-time electronic health record data (e.g., labs, vital signs, history, imaging). Participants must consent to data collection.

Description

Inclusion Criteria:

1. Patients admitted to any department of the hospital (e.g., ICU, general wards, emergency, outpatient services) during the study period.
2. Patients with available real-time electronic health record (EHR) data, including at least two of the following: laboratory results, vital signs, medical history, and imaging data.

Exclusion Criteria:

Patients currently enrolled in another clinical trial that could interfere with data collection or outcomes of this study.

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort	Intervention / Treatment
Hospital-Wide Patient Cohort	Other: AI-associated strategy; The intervention in this study involves an AI system that leverages multimodal data fusion to support the clinical decision-making and evaluation of diseases. Patients in this cohort will undergo standard examinations, with clinical decisions guided by the recommendations generated by the AI system.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Area Under the Curve (AUC)	AUC of the ROC curve, used to quantify diagnostic accuracy. No unit (a ratio or percentage, typically expressed as a number between 0 and 1).	1 year
Overall Hospital Resource Utilization Improvement	The percentage reduction in overall hospital resource use (e.g., bed days, ICU admissions, diagnostic tests) attributed to AI-assisted decision-making, expressed as a percentage.	1 year
Population-Level Diagnostic Accuracy Enhancement	The overall improvement in diagnostic accuracy across all hospital patients (e.g., percentage of correct diagnoses or reduction in misdiagnoses) facilitated by the AI model, expressed as a percentage or ratio.	1 year
System-Wide Reduction in Adverse Event Rates	The percentage reduction in major adverse events (e.g., mortality, severe complications, or prolonged stays) across all hospital patients due to AI-assisted decision-making, expressed as a percentage.	1 year

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Overall Improvement in Hospital Patient Outcomes	The aggregate improvement in key patient outcomes (e.g., mortality, morbidity, recovery rates) across the entire hospital population due to AI-assisted decision-making, expressed as a composite score or percentage.	1 year
Enhancement of Healthcare System Efficiency	he overall improvement in hospital operational efficiency (e.g., reduced wait times, optimized resource allocation, decreased staff workload) attributed to the AI model, expressed as a percentage or qualitative rating.	1 year
Population Health Impact Score	A composite score reflecting the AI model's effect on population health within the hospital's catchment area (e.g., reduced disease burden, improved chronic disease management), expressed as a standardized index or percentage change.	1 year
Long-Term Public Health Benefit Index	A composite index measuring the AI model's long-term contribution to public health (e.g., reduced disease prevalence, improved life expectancy), expressed as a standardized score or percentage improvement.	1 year

Collaborators and Investigators

• Sponsor: The Eye Hospital of Wenzhou Medical University

Study record dates

Study Major Dates

• Study Start (Actual): February 24, 2025
• Primary Completion (Estimated): June 1, 2026
• Study Completion (Estimated): June 1, 2026

Study Registration Dates

• First Submitted: February 24, 2025
• First Submitted That Met QC Criteria: February 24, 2025
• First Posted (Actual): March 25, 2025

Study Record Updates

• Last Update Posted (Actual): March 25, 2025
• Last Update Submitted That Met QC Criteria: February 27, 2025
• Last Verified: February 1, 2025

More Information

Terms related to this study

• Keywords: outcome; diagnosis; prediction; AI
• Other Study ID Numbers: AI Prediction

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No
• product manufactured in and exported from the U.S.: No