New medical software that tracks heart disease in individual patients brought closer to roll-out
State-of-the-art software that can model and predict how diseases behave in an individual patient may be part of the future of medicine, but implementing it in hospitals remains a challenge. Hampered by a lack of testing in real clinical environments as well as the complexity of the programmes for healthcare professionals, cutting edge medical technology can find it hard to become established. EU-funded project CARDIOPROOF (Proof of Concept of Model-based Cardiovascular Prediction) has helped to bridge the gap between the successful development of new computer-modelling technology and its deployment in a hospital or clinic. ‘Our project helped to boost confidence in predictive computer models, as well as their acceptance by healthcare professionals,’ says Edwin Morley-Fletcher, CARDIOPROOF project coordinator. The project focussed on aortic valve disease and aortic coarctation in paediatric patients which, if left untreated, can result in irreversible heart failure. Researchers tested computer programmes that can help achieve an early diagnosis of the diseases, predict disease behaviour and evolution, and predict treatment outcomes. CARDIOPROOF tested four tools: advanced modelling solutions for heart electro-mechanics, modelling of the blood flow through the aortic arch using the Smooth Particle Hydrodynamic approach, a Fluid-Structure interaction model and modelling of the aortic coarctation simulating the relevant treatment options. It sought to validate these tools in clinical trials held at three European centres of excellence in cardiac treatment: The Ospedale Pediatrico Bambin Gesù in Rome, University College London’s Great Ormond Street Hospital for Children NHS Foundation Trust (GOSH), and the Deutsches Herzzentrum Hospital in Berlin. CARDIOPROOF developed a randomised experiment to assess the impact of the modelling tools within clinical decision making. It involved 172 cardiologists who were guided through three different decision-making scenarios and were divided into two groups. One group received data currently used to determine treatment paths, and the other group received model-based data, in addition to ‘normal’ data. ‘This experiment showed the potential positive impact of being supported by computer-modelling in clinical decision-making. By providing useful insight to choose the best treatment options and the correct timing for the intervention we have encouraged further research in this field,’ says Morley-Fletcher. In addition, CARDIOPROOF boosted the evidence of the clinical benefits of using computer-based models for a virtual stenting tool which assesses optimal treatment options. It also showed the benefits of a computational pressure mapping tool which helps avoid risky and invasive use of catheters to detect blood pressure. The project also worked on ways to improve the user-friendliness of the computational tools and ways to reduce relevant hardware requirements. It also developed web-based tools accessible from a normal internet browser, without any installation required, greatly enhancing usability in clinical settings. Since the end of the project in December 2016, some of the tools tested have been integrated in commercial software. Meanwhile, the datasets and the data management platform developed during the CARDOPROOF project will be used in a new EU-Funded project, MYHEALTHMYDATA. This project aims to create a computer-based architecture for secure patient data storage, management and exchange.
Keywords
CARDIOPROOF, heart disease, clinical testing, computational modelling, treatment, paediatrics
