Use of Augmented Intelligence for the Interpretation of Bone Standard X-rays Prescribed by the Emergency Department (IMMEDIAT Urgences)

Assistance Publique - Hôpitaux de Paris

Status: Enrolling

Conditions: Bone Fracture

Treatments: Other: Organization with AI

Study type: Interventional

Funder types: Other

Identifiers

NCT05882435

21OLE

Details and patient eligibility

About

The study evaluates, in 3 emergency departments (ED) and on randomized alternate periods, the use of SmartUrgences®, Augmented Intelligence (AI) software to help the interpretation of bone standard X-rays prescibed by the ED.

Full description

"The study evaluates, in 3 emergency departments (ED) and on randomized alternate periods, the use of SmartUrgences®, Augmented Intelligence (AI) software to help the interpretation of bone radiographs prescibed by the ED.

The primary objective of the organizational study is to evaluate, compared to the current organization for the interpretation of standard bone radiographs requested by the ED, the impact of an organization incorporating the Milvue solution, on the reduction of the patient diagnostic error rate.

A cost-consequence study is carried out, comparing from the point of view of the community (production costs according to the HAS recommendations), the radiological diagnosis within the framework of an organization with the Milvue solution, to that within the framework of the current organization without the use of the Milvue solution.

The economic study will follow the scheme of the organizational study, comparing the periods with and without the Milvue solution and analyzing the costs and consequences, by period and by patient.

This is an open-label randomized cluster multiple period cross-over study with 6 alternate periods (3 with AI, 3 with usual organization) of 1 month in each ED. The choice of the intervention for the first period will be randomized."

Enrollment: 8,400 estimated patients

Sex: All

Ages: 18+ years old

Volunteers: No Healthy Volunteers

Inclusion and exclusion criteria

Inclusion Criteria:

1. All adult referred by the ED for a conventional X-rays of all or part of the appendicular skeleton and/or pelvis and/or costal gril
2. Not opposed to participate"

Trial design

Primary purpose: Health Services Research

Allocation: Randomized

Interventional model: Crossover Assignment

Masking: None (Open label)

---

8,400 participants in 2 patient groups

Usual Organization

No Intervention group

Description:

X-rays are done in the radiology department and the images are made available to emergency physicians without waiting for the radiologist's report, which is usually done on a delayed basis.

Organization with AI

Other group

Description:

X-rays are done in the radiology department and the images are made available to emergency physicians with the AI interpretation. X-rays flagged by IA as anormal or suspicious will be reviewed without delay by the radiologist, non-flagged X-rays will be reviewed by radiologists on a delayed basis.

Treatment:

Other: Organization with AI