NCT06118840

IDEAL Study: Blinded RCT for the Impact of AI Model for Cerebral Aneurysms Detection on Patients' Diagnosis and Outcomes

Assessing the Impact of an Artificial Intelligence-Based Model for Intracranial Aneurysm Detection in CT Angiography on Patients' Diagnosis and Outcomes: The IDEAL Study - A Web-Based Multicenter, Double-Blinded Randomized Controlled Trial

• Status: Recruiting
• Phase: N/A
• Study type: Interventional
• Enrollment: 6,450 (estimated)
• Sponsor: Jinling Hospital, China · Academic / Other
• Sex: All
• Age: 18 Years
• Healthy volunteers: Accepted

Summary

This study (IEDAL study) intends to prospectively enroll more than 6450 patients who will undergo head CT angiography (CTA) scanning in the outpatient clinic. It will be carried out in 21 hospitals in more than 10 provinces in China. The patient's head CTA images will be randomly assigned to the True-AI and Sham-AI group with a ratio of 1:1, and the patients and radiologists are unaware of the allocation. The primary outcomes are sensitivity and specificity of detecting intracranial aneurysms. The secondary outcomes focus on the prognosis and outcomes of the patients.

Detailed description

A multicenter, prospective, double-blind, randomized controlled trial will be conducted (IDEAL study). Patients who are scheduled to undergo cranial CT angiography (CTA) scanning will be randomly divided into two groups with a ratio of 1:1, one of the group will be assigned to True-AI aided intracranial aneurysms diagnosis strategy (True-AI group) and the other will be assigned to Sham-AI aided intracranial aneurysms diagnosis strategy (Sham-AI group, which has a sensitivity close to 0% and a similar specificity to True-AI). The primary outcomes are diagnostic sensitivity and specificity of detecting aneurysms. Secondary endpoints include other diagnostic performance indexes for intracranial aneurysms; diagnostic performances for other intracranial lesions for intracranial arterial stenosis, occlusion, and intracranial tumors; detection rates of intracranial lesions according to Radiology Reports; workload of head CTA interpretation; resource use; treatment-related indexes during patient follow-up (e.g. clinical follow-up, hospitalization, rate of patients undergoing DSA); life quality; outcomes of aneurysm-related events; repeat head CTA or MRA at 12-month follow; cost-effectiveness analysis between intervention and control arm to evaluate the short- and longterm influence of AI system to the routine practice and patients' prognosis and outcomes.

Conditions

• Intracranial Aneurysm
• CT Angiography
• Deep Learning
• Double Bind Interaction

Interventions

Type	Name	Description
DEVICE	True-AI-integrated intracranial aneurysms diagnosis strategy	The True-AI deep-learning based model for intracranial aneurysms detection had a patient-wise sensitivity, lesion-wise sensitivity and specificity of 0.96, 0.87, and 0.80 in the internal validation dataset.
DEVICE	Sham-AI-integrated intracranial aneurysms diagnosis strategy	The Sham-AI deep-learning based model for intracranial aneurysms detection is designed to have a sensitivity close to 0% and a similar specificity to the True-AI. In the internal validation dataset, the Sham-AI had a patient-wise sensitivity, lesion-wise sensitivity, specificity of 0.02, 0.01, and 0.80, respectively.

Timeline

• Start date: 2024-05-20
• Primary completion: 2026-07-01
• Completion: 2026-12-01
• First posted: 2023-11-07
• Last updated: 2025-10-07

Locations

21 sites across 1 country: China