Real-world Effectiveness Evaluation of Clinical Decision Support System Based on Artificial Intelligence (AI-CDSS)

Real-world Effectiveness Evaluation of Clinical Decision Support System Based on Artificial Intelligence (AI-CDSS) on Diagnosis

This study intends to explore the accuracy of clinical diagnosis of AI based CDSS system and promotion of clinical work by comparing CDSS before and after the online.

Study Overview

• Status: Completed
• Conditions: Medical Informatics Applications
• Intervention / Treatment: Other: Auxiliary diagnostic system by AI-CDSS

Detailed Description

This study intends to explore the accuracy of clinical diagnosis of AI based CDSS system and promotion of clinical work by comparing CDSS before and after the online. The difference of diagnostic accuracy before and after AI-CDSS application will be compared by the before and after design, and the role of AI-CDSS will be explored.

• Study Type: Observational
• Enrollment (Actual): 34113

Contacts and Locations

Study Locations

• China
  • Beijing, China
    • Peking University Third Hospital

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Child; Adult; Older Adult
• Accepts Healthy Volunteers: No
• Genders Eligible for Study: All

• Sampling Method: Non-Probability Sample

Study Population

Study population was the hospitalized patient from December 2016 to February 2019 in 6 clinical departments. The six clinical departments were Otolaryngology, Orthopaedic, Respiratory Medicine, General Surgery, Cardiology and Hematology.

Description

Inclusion Criteria:

• all hospitalized patients in 6 clinical departments, Otolaryngology, Orthopaedic, Respiratory Medicine, General Surgery, Cardiology and Hematology from December 2016 to February 2019.

Exclusion Criteria:

• Missing data for key variables

Study Plan

How is the study designed?

Number of groups / cohorts

2

Cohorts and Interventions

Group / Cohort	Intervention / Treatment
Before; before CDSS on-line
After; after CDSS on-line	Other: Auxiliary diagnostic system by AI-CDSS; Helping clinicians to make diagnoses by using CDSS based-on AI

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Accuracy Rate of Recommended Diagnosis by CDSS, up to 12 weeks	Based on the patient's discharge diagnosis as a standard, it is explored whether the diagnosis given by the CDSS is consistent with the discharge diagnosis in the patient's medical record.	When the subject was discharged from the hospital
Patients' hospitalization time (days), up to 24 weeks	The length of a patient's stay is the number of days he or she experiences from the time of admission to the time of discharge.	When the subject was discharged from the hospital
consistency between admission diagnosis and discharge diagnosis up to 12 weeks	When the patient comes to the hospital, the clinician will write an inpatient record and give a preliminary diagnosis, which we call admission diagnosis.After the patient is hospitalized, all kinds of examinations will be improved. After all the examination results come out, the patient's diagnosis on admission may be modified. Because there are no auxiliary examination results on admission, the diagnosis on admission may not be completely correct.This modified diagnosis is called discharge diagnosis.This study compared the consistent rate of admission diagnosis and discharge diagnosis before and after CDSS on-line.	When the subject was discharged from the hospital

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
length of confirmed time, up to 6 weeks	he length of confirmed time (days) was the duration between the preliminary admission diagnosis and the definite diagnosis.	When the subject was discharged from the hospital

Collaborators and Investigators

• Sponsor: Peking University Third Hospital
• Investigators: Principal Investigator: Hong Qi, Ph.D, Peking University Third Hospital

Study record dates

Study Major Dates

• Study Start (Actual): March 1, 2019
• Primary Completion (Actual): May 1, 2019
• Study Completion (Actual): June 1, 2019

Study Registration Dates

• First Submitted: July 12, 2019
• First Submitted That Met QC Criteria: September 23, 2021
• First Posted (Actual): October 4, 2021

Study Record Updates

• Last Update Posted (Actual): October 4, 2021
• Last Update Submitted That Met QC Criteria: September 23, 2021
• Last Verified: July 1, 2019

More Information

Terms related to this study

• Other Study ID Numbers: 324-01

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: UNDECIDED
• IPD Plan Description: After the results of the study are published for one year, the data can be published to other researchers. But the data needs to be de-identified to protect the patient's privacy.

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No