An Observational Clinical Study on the Construction of an Artificial Neural Network Model for ICU Pneumonia

Not yet recruiting

• Conditions: Pneumonia

• Registration Number: NCT06661499
• Lead Sponsor: Chinese Medical Association

Brief Summary

To achieve rapid, intelligent and accurate microbiological diagnosis and treatment for ICU pneumonia, an artificial neural network model for microbiological diagnosis is established, which depends on many clinical cases and machine deep learning from clinical experts' judgements according to species-specific rapid detection of pathogenic bacteria and other clinical parameter variables of patients.

Detailed Description

This study is a prospective single-centre observational study, 600 ICU pneumonia patients are expected to be selected as the observation object, and the lower respiratory secretions of patients on d1, d3 and d7 after enrollment are collected for species-specific rapid detection and microbial culture, while the general information of the patients and the clinical information of the corresponding time points on d1, d3 and d7 are collected. Two experienced senior physicians were organized to determine whether the microbial results were colonized or infected, and an artificial neural network model for rapid and intelligent diagnosis of pathogenic microorganisms in ICU pneumonia will be established and validated through multi-dimensional machine learning.

Study Timeline

• Status Verified: 2024-08
• Study First Submitted: 2024-09-05
• Study First Submitted QC: 2024-10-24
• Last Update Submitted: 2024-10-24
• Study First Posted: 2024-10-28
• Last Update Posted: 2024-10-28
• Study Start: 2025-01
• Primary Completion: 2027-12
• Study Completion: 2027-12

Recruitment & Eligibility

• Status: NOT_YET_RECRUITING
• Sex: All
• Target Recruitment: 600

Inclusion Criteria

1. aged ≥18 years;
2. agreed to obtain lower respiratory specimens for rapid testing of pathogenic bacteria;
3. all were enrolled by an experienced physician who dynamically determined that the microorganisms were in a colonised or infected state;
4. signed an informed consent form.

Exclusion Criteria

1. pregnant women;
2. lactating women;
3. patients who could not obtain lower respiratory specimens;

Study & Design

• Study Type: OBSERVATIONAL
• Study Design: Not specified

Primary Outcome Measures

Name	Time	Method
clinical evaluation of each microbial detected whether in colonization or infection	day1,day3 and day7 after enrollment	Two experienced senior physicians are organized to determine whether the microbial are colonized or infected according to clinical values.

Trial Locations

Locations (1)

Nanjing Drum Tower Hospital; 🇨🇳 Nanjing, Jiangsu, China