NCT04394052

Novel Imaging Techniques for the Characterization of Musculoskeletal Tumors II

Novel Imaging Techniques for the Characterization of Musculoskeletal Tumors II: Texture Analysis and Artificial Intelligence

• Status: Recruiting
• Phase: —
• Study type: Observational
• Enrollment: 740 (estimated)
• Sponsor: Central Hospital, Nancy, France · Academic / Other
• Sex: All
• Age: 18 Years
• Healthy volunteers: Not accepted

Summary

This study aims at evaluating the value of various artificial intelligence based techniques to improve the characterization and image post-processing for patients with musculoskeletal tumors.

Detailed description

Comparison of values relating to the texture parameters of tumors evaluated by MRI and ultra-high resolution CT between benign and malignant lesions using histological analysis as the standard of reference. Comparison of the diagnostic performance of texture parameters derived from different MRI sequences and ultra-high resolution CT for musculoskeletal tumor characterization. Evaluate the impact of ultra-high resolution with respect to standard resolution on CT images Comparison of the diagnostic performance of the texture parameters for the tumor on the diagnostic performance of texture analysis derived parameters for the characterization of musculoskeletal tumors. Evaluate the effectiveness and accuracy of automatic artificial intelligence (AI) based tumor segmentation tools. Evaluate the use of trabecular analysis on ultra-high resolution CT images for the evaluation of tumor-bone interfaces.

Conditions

• Soft Tissue Neoplasm
• Bone Neoplasm

Interventions

Type	Name	Description
DIAGNOSTIC_TEST	MR imaging	Medical imaging

Timeline

• Start date: 2020-06-01
• Primary completion: 2026-06-01
• Completion: 2030-06-01
• First posted: 2020-05-19
• Last updated: 2020-07-21

Locations

1 site across 1 country: France