Novel Patient Stratification based on Photon Counting CT

Each year, 3.9 million lives are lost to cardiovascular diseases (CVD) within Europe and 1.8 million within the European Union. 45% of all deaths within Europe are due to CVD, and in 2015, more than 85 million people in Europe are living with CVD. Furthermore, CVDs are arguably the most important comorbidities in chronic obtrusive pulmonary disease (COPD) as COPDs are a common presence in patients with CVD, and they are often associated with increased risk for hospitalisation, longer stay in hospitals and CVD-related mortality . Of the total cost of CVD to the EU, 53% (€111 billion) is due to health care costs, 26% (€54 billion) is due to productivity losses and 21% (€45 billion) to the informal care of people living with CVDs. Computational approaches utilizing big and complex data hold great potential to enable superior patient stratification, with the potential to significantly improve outcomes of CVD and COPD. In OpenThorax, we seek to redefine the current SOA patient stratification methods for CVDs, COPDs and COVID-19 with a non-invasive thoracic imaging workflow, backed by new Photon Counting CT (PCCT) imaging coupled with radiomics, a novel approach to physics-informed machine learning, and Digital Twinning. These will combine to form an integrated clinical software united by a secure cloud-based clinical knowledgebase system. Two layers of stratification are used; one based on empirical values from Photon Counting CT (PCCT) images, and a secondary stratification based on multi-physics computations & physics-informed AI predictive & prognosis models. The first stratification layer aims to reduce Time to Diagnosis (TTD) and time for symptom resolution, while the second layer aims to improve outpatient care and outcome.