Automatic GPU-based Algorithm for OMERACT-RAMRIS Score Quantification in Rheumatoid Arthritis

Abstract

Rheumatoid Arthritis (RA) is a systemic disease that affects the synovial joints. Currently, the gold standard measurement for tracking the progression of the disease involves a semi-quantitative assessment of bone erosion, bone marrow edema, and synovitis in Magnetic Resonance Images. The Outcome Measures in Rheumatology Clinical Trials (OMERACT) Rheumatoid Arthritis Magnetic Resonance Imaging Score (RAMRIS) system represents a useful standard for the assessment of RA. However, the use of RAMRIS system is time-consuming and tedious and requires a long learning curve. The work presented in this paper is aimed at identifying how computer automation, using parallel computing on the Graphics Processing Unit (GPU), can be used to quantify OMERACT-RAMRIS score. The proposed algorithm is fully integrated in a Computer Aided Diagnosis (CAD) system named RheumaSCORE (Softeco Sismat Srl) and enables users-even non-expert ones-to evaluate bone erosion in a short time and with little training. Preliminary results of qualitative and quantitative validation are presented and discussed at the end of the paper.