IA-IRM-SIJ-SPA

Automatic detection of Axial Ankylosing Spondylitis on sacroiliac MRI via Deep Learning

Underway

02/05/2023

02/05/2027

Main objectives

Early diagnosis of Ankylosing Spondylitis (SpA) is essential for optimal treatment, and requires a number of criteria.
The project aims to develop and validate an artificial intelligence algorithm capable of detecting sacroiliac joint abnormalities suggestive of axial spondyloarthritis on MRI of the pelvis.

#DeepLearning | #MRI | #SacroIliac | #Spondyloarthrite

Challenge

Spondyloarthritis is the most common inflammatory rheumatism in young adults (0.2 to 0.4% of the population). The main symptoms of this rheumatic disease are pain in the spine and buttocks, of an inflammatory type (responsible for waking up at night and waking up in the morning). It causes considerable disability at first, due to inflammation of the pelvic joints (sacroiliac) and spinal ligaments, then progressively due to ossification of the vertebral ligaments, which results in poor posture (ankylosis and kyphosis) and ventilatory problems. This progression can be prevented by early diagnosis and appropriate treatment, particularly since the advent of biotherapies in recent decades.

Methodology/Technology used

Deep learning methods

Contacts

Antoine Feydy () & Hugues Talbot ()& Théodore Aouad ()

Members

Inria, Centrale-Supelec, AP-HP