Transforming ED Throughput With AI-Driven Clinical Decision Support System

National Taiwan University Hospital4,016 enrolled

Overview

The aims of this study is to integrate real-time data flow infrastructure between hospital information system and AI models and to conduct a cluster randomized crossover trial to evaluate the efficacy of the AI models in improving patient flow and relieving ED crowding.

Study Type

INTERVENTIONAL

Allocation

RANDOMIZED

Purpose

HEALTH_SERVICES_RESEARCH

Masking

NONE

Enrollment

4,016

Conditions

Critical Care Emergency Treatment Triage Readmission

Interventions

AI-assisted models providing diagnosis and prognostic informationOTHER

AI-assisted models providing diagnosis and prognostic information in the ED, including triage, ICD coding, chest x ray alerts, critical event alerts, readmission prediction, and post-cardiac arrest prognostication.

Critical treatmentPROCEDURE

Critical treatment of the emergency room

Eligibility

Sex: ALLMin age: 20 Years

Medical Language ↔ Plain English

Inclusion Criteria: * ED patients aged 20 years or older * Patients were treated by the recruited 16 ED attendings. Exclusion Criteria: * Patients aged less than 20 years. * Patients were not treated by the recruited 16 ED attendings.

Locations (1)

National Taiwan University Hospital

Taipei, Taiwan

Outcomes

Primary Outcomes

ED length of stay

Time frame: From ED arrival to 3 days after ED discharge. For hospitalized patients with cardiac arrest, the outcome ascertainment continues until hospital discharge.

Data from ClinicalTrials.gov

This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.