AutoPiX

Patients at risk or diagnosed with arthritis are constantly assessed by innovative imaging techniques to document the onset or progression of their disease. However, despite their impressive abundance and resolution, these images lack the analysis and interpretation tools necessary to deliver unbiased and precise diagnosis, monitoring and prognosis to the patients. Additionally, some key advanced imaging methodologies such as ultrasound are hardly accessible to most of patients, urging improvements of more accessible imaging methods. The AutoPiX project is an ambitious international multi-stakeholder effort led jointly and synergistically by academic and industry partners to achieve breakthroughs in both the applicability and harnessing of imaging technologies for the benefit of patients by creating new powerful analysis and decision tools. We will first generate tools to transform unstructured images into quantitative biomarkers using artificial intelligence (AI) and machine learning (ML) models, and validate them clinically for their diagnosis, monitoring and prognosis power. This will considerably increase the utility of imaging biomarkers for arthritis and bring them to the same level as laboratory biomarkers. In parallel we will develop accessible imaging strategies such as remote monitoring and robotic-powered point-of-care ultrasound exams for patients to mitigate the often-observed shortage of qualified personnel in real world settings. To achieve this, we will improve the precision and interpretability of these methods and further validate them with clinical, molecular and histological analyses. Our consortium is built on multi-disciplinarity and the constant synergistic interaction of all the actors of arthritis care: rheumatologists, radiologists, patients, researchers, regulators, industries and small- and medium sized enterprises (SMEs). On the long term, AutoPiX will create new clinically-validated methods to achieve a/ more precise, accessible and effective diagnosis, b/ shortened and better-tailored treatment paths and c/ improved treatment response assessments and outcome prediction for patients with arthritis.