An Algorithm Creation by Automated Segmentation by MRI for Bone and Muscles of Shoulder

Not Applicable

Withdrawn

• Conditions: Reverse Shoulder Prosthesis
• Interventions: Procedure: CT-Scan and MRI

• Registration Number: NCT05376813
• Lead Sponsor: Assistance Publique - Hôpitaux de Paris

Brief Summary

The primary objective of the study is to develop an algorithm of automated segmentation of shoulder by MRI examinations.

Detailed Description

This is a national monocentric study which will be conducted in Ambroise Paré hospital of APHP, in orthopaedics department (for enrollment) and radiological department (for CT-scan and MRI examinations) respectively.

Manual segmentations of 5 muscles and 2 bones of shoulder by MRI with automated segmentation of shoulders corresponding to CT-scan imagings.

3D imagings of each shoulder by manual segmentations from MRI and automated segmentation from computed tomography will provide to build a network.

The perspective of the elaborated algorithm should lead to an automated 3D-reconstruction of patients' shoulder as a routine care in surgery.

Study Timeline

• Study First Submitted: 2022-05-09
• Study First Submitted QC: 2022-05-16
• Study First Posted: 2022-05-17
• Study Start: 2023-08-31
• Primary Completion: 2024-04-28
• Study Completion: 2024-04-28
• Status Verified: 2024-06
• Last Update Submitted: 2025-03-17
• Last Update Posted: 2025-03-19

Recruitment & Eligibility

• Status: WITHDRAWN
• Sex: All
• Target Recruitment: 100

Inclusion Criteria

• Healthy volunteer > 18 years, presenting any symptom nor history of shoulder pathology;
• Affiliated to social security scheme.

Exclusion Criteria

• Symptoms or history of shoulder pathologies;
• Claustrophobia;
• Pregnant woman;
• Patient covered by AME system;
• Contre-indication to perform MRI examination (implant, less 6-months stent implantation, recent surgery, renal insufficiency, pace maker implantation, cardiac defibrillator, cardiovascular catheter, neurostimulation, implantable electronic pompe for automatic injection of medications).

Study & Design

• Study Type: INTERVENTIONAL
• Study Design: SINGLE_GROUP

Arm && Interventions

Group	Intervention	Description
Experimental arm	CT-Scan and MRI	All participants are healthy volunteers

Primary Outcome Measures

Name	Time	Method
Algorithm developement	through study completion, an average of 8 month	The developement for automatic segmentation algorithm: uses method with a convolutional neural networks (convolutional neural network - CNN).

Trial Locations

Locations (1)

Orthopaedics, Ambroise Paré hospital, APHP; 🇫🇷 Boulogne-Billancourt, France