Mitral valve regurgitation is the most common valvular disease, affecting 10% of the population over 75 years old. Current standard of care diagnostic imaging for mitral valve procedures primarily consists of transesophageal echocardiography (TEE) as it provides a clear view of the mitral valve leaflets and surrounding tissue. Patient-specific physical valve models for functional simulation based on 3D printable molds have been developed with applications in surgical training and procedure planning, however, producing a mesh model of the valve geometry from TEE imaging remains a challenge. TEE has limitations in signal dropout and artefacts, particularly for structures lying below the valve such as chordae tendineae. We have developed a volume compounding system to fuse multiple TEE acquisitions to create a single volume containing the mitral valve and sub-valvular structures with a high level of detail. Capturing the necessary additional volumes can be done with no added cost and only an additional ten minutes to the current standard of care diagnostic images. This enables us to enhance the utility of existing cardiac imaging systems through an image processing-based approach. Virtual reality-based platforms for image visualization enable direct 3D interaction with complex volumetric ultrasound data, as well as potentially offering remote multi-user interaction. We have also developed DeepMitral to enable fully automated mitral valve leaflet segmentations. From leaflet segmentations, we can extract the atrial surface and generate a positive mold of the patient-specific leaflet geometry. These molds can then be 3D printed to be used in the production of dynamic silicone valve models for functional simulation in our beating heart phantom, enabling various repair techniques to be evaluated prior to surgery. The application of image processing, virtual reality visualization, and physical functional simulation to mitral valve procedures altogether serves to enable clinicians to better plan their repair strategy.