Developing a machine learning tool to improve prognostic and treatment response assessment on cardiac MRI data (360G-Wellcome-215799_Z_19_Z)

This funding will support adaption/evaluation of a novel tensor-based machine learning tool for CMRI image analysis focusing on pulmonary hypertension. This application is to enable: - Assembly and management of cohorts and data - Assemble CMRI from one prospective cohort and one large registry cohort - Assess image quality - Assemble health record and treatment data - Adaption of our technology to make prognostic and treatment response assessments - Adapt our developed CMRI diagnosis tool for prognostic treatment response prediction - Adapt our developed CMRI diagnostic visualisation tool for disease responsive pattern interpretation - Integrate health record data with CMRI for prediction and interpretation - Evaluation - Perform repeatability and treatment response assessment using prospective cohort - Perform baseline and follow-up prognostic accuracy assessment on large registry cohort - Evaluate the improvement in prediction by integrating health record data - Develop the prototype and engage industry partners to facilitate commercialisation Wider vision: - Clinical utilisation of our technology for diagnosis, prognostication and treatment response assessments to achieve an overarching ambition in disease assessment. - Adaption and application of our technology to: other modalities such as Echocardiography and computed tomography, other cardiac diseases and for use in drug discovery.