AI-Driven Prediction of Hospital-Acquired Infections With EHR

NCT06791382 | Recruiting

• 1 other identifier: Hospital-Acquired Infections
• Study Type: observational
• Target: 1,000,000
• Locations: 1 country
• Sites: 2

Brief Summary

This is a multi-center, clinical study designed to evaluate the application and effectiveness of an AI-assisted predictive model for identifying and diagnosing infection, leveraging multimodal health data.

Conditions

Hospital-acquired Infections

Outcome Measures

Primary Outcomes (2)

• 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

• F1 Score

  The F1 score is the harmonic mean of precision and sensitivity (recall). It is a good measure of the model's ability to identify both true positives and minimize false positives, especially in cases where the classes are imbalanced (e.g., when the number of healthy cases is much higher than disease cases). The F1 score ranges from 0 to 1, with 1 indicating perfect precision and recall.

  1 year

Secondary Outcomes (2)

• Sensitivity (True Positive Rate)

  1 year

• Specificity (True Negative Rate)

  1 year

Study Arms (2)

Hospital-Acquired Infection Cohort

This group consists of patients who have developed a hospital-acquired infection (HAI) during their hospital stay. Participants in this cohort will be used to evaluate the effectiveness of the AI-assisted predictive model in identifying the risk factors leading to hospital-acquired infections. The model will be assessed based on the accuracy of predicting infection risks in hospitalized patients. No specific interventions will be provided as part of this cohort beyond the existing hospital infection control practices.

Diagnostic Test: AI-Based Diagnostic and Prognostic Model

Healthy Cohort (No HAI)

This group consists of patients who have not developed any hospital-acquired infections during their hospital stay. Participants in this cohort will serve as the control group for comparison against the experimental group. The AI-assisted model will be evaluated for its ability to distinguish between patients who are at risk for developing infections and those who remain infection-free during hospitalization. No interventions will be provided as part of this cohort, as they represent patients without infection-related complications.

Diagnostic Test: AI-Based Diagnostic and Prognostic Model

Interventions

AI-Based Diagnostic and Prognostic Model DIAGNOSTIC_TEST

This intervention involves an AI system that integrates multimodal data, including patient medical history, laboratory test results, clinical observations, and treatment data, to predict the risk of hospital-acquired infections (HAIs). The system uses deep learning algorithms to provide real-time, accurate predictions, enabling early identification of patients at risk for infections. By analyzing historical health data, the model aims to predict potential infection developments, improving early detection, prevention strategies, and patient outcomes in hospital settings.

Healthy Cohort (No HAI); Hospital-Acquired Infection Cohort

Eligibility Criteria

• Age: 0 Years - 90 Years
• Sex: all
• Healthy Volunteers: Yes
• Age Groups: Child (0-17), Adult (18-64), Older Adult (65+)
• Sampling Method: Non-Probability Sample

Study Population

The study population consists of individuals who have received care at participating study centers. Participants must have comprehensive electronic health records (EHRs) available, including medical history, laboratory test results, treatment data, and clinical observations. Both individuals who have developed hospital-acquired infections (HAIs) and those who have not will be included in the study to evaluate the AI-assisted model's predictive capabilities for infection risk. The study will focus on patients with complete and documented care records from the participating centers, ensuring a diverse cohort for analysis across different age groups and infection types.

You may qualify if:

• Patients with complete and accessible EHR data, including medical history, laboratory test results, treatment regimens, clinical observations, and infection history.
• Patients who have been admitted to the participating hospital or healthcare facility during the study period.
• All participants must provide informed consent to use their health data for research purposes.

You may not qualify if:

• Patients with incomplete or missing critical EHR data, such as lab results, medical history, or treatment details, which are necessary for infection prediction.
• Patients who have severe cognitive disorders, dementia, or conditions that prevent them from providing informed consent or participating in the study.
• Patients who have not been admitted to the hospital during the study period or who are receiving outpatient care only.
• Patients with terminal conditions where infection prediction may not be applicable to the clinical goals of the study.

Sponsors & Collaborators

• The Eye Hospital of Wenzhou Medical University: lead

Study Sites (2)

First Affiliated Hospital of Wenzhou Medical University

Wenzhou, Zhejiang, China

RECRUITING

Second Affiliated Hospital of Wenzhou Medical University

Wenzhou, Zhejiang, China

RECRUITING

Central Study Contacts

Fei Liu, MD

CONTACT

+86 13810512704; liufei_2359@163.com

Study Design

• Study Type: observational
• Observational Model: CASE CONTROL
• Time Perspective: OTHER
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Chief Scientist

Study Record Dates

• First Submitted: January 19, 2025
• First Posted: January 24, 2025
• Study Start: February 1, 2023
• Primary Completion: May 1, 2025
• Study Completion: May 1, 2025
• Last Updated: April 17, 2025

Record last verified: 2025-04

Data Sharing

• IPD Sharing: Will not share

Locations

CN (2)