ZusammenfassungMagnetic resonance imaging (MRI) of the sacroiliac (SI) joints is utilized for the assessment of active inflammatory and structural damage in patients with suspected axial spondyloarthritis (axSpA) for accurate diagnosis and correct classification. The purpose of the study is to develop an algorithm that is able to assess the presence of active and chronic inflammatory lesions typical for axSpA on MRI of the SI joints by using machine learning, a form of artificial intelligence. We are planning to use German National Cohort Study (GNC) MRI data of approximately 2,000 participants which will be assessed by three readers with experience in SpA to train the algorithm and to assess the accuracy of the trained algorithm. The developed algorithm will be applied to the remaining GNC MRI data (not used for training and validation) available for automated detection of changes compatible with axSpA. We expect that the developed algorithm will be a useful tool in clinical practice providing help in the interpretation of MRIs of the SI joints in patients with suspicion of axSpA, supporting diagnostic decision and simplifying the process of classification.

Schlüsselwörter-

EinrichtungenCharité – Universitätsmedizin Berlin, Universitätsmedizin Greifswald