Towards smarter cardiac interventional navigation using MRI-derived personalized models for RT 3D ultrasound interpretation

     Minimally-invasive cardiovascular interventions are performed daily to treat several CVDs, including coronary heart diseases, cardiomyopathies, arrhythmias, valvular heart diseases, among others. The success of these procedures is directly linked to the physician’s ability to precisely visualize and target the surgical site. To this end, several imaging modalities (e.g. X-ray fluoroscopy, echocardiography, magnetic resonance imaging - MRI, etc.) are presently being used for interventional guidance. 
     In the current project, we propose a novel concept for smarter and more accurate image guided surgeries, where the preoperative data is used to generate personalized models that will enhance the information extracted from the interventional images. Additionally, shape-based non-rigid registration between the interventional and pre-operative images will be performed, enabling accurate image fusion.

MRI

Echo

Additional Information

  • FCT - EXPL/BBB-BMD/2473/2013.
  • Daniel Barbosa, Sandro Queirós, Jan D'hooge, João L. Vilaça (2015). Improving the robustness of interventional 4D ultrasound segmentation through the use of personalized prior shape models. SPIE Medical Imaging 2015, 21 - 26 February 2015, Orlando, United States.
  • Andrea Perissinotto, P. Morais, S. Queirós, N. F. Rodrigues, M. J. Baptista, J. D'hooge, João. L. Vilaça, D. Barbosa (2015). Robust temporal alignment of multimodal cardiac sequences, SPIE Medical Imaging 2015, 21 - 26 February 2015, Orlando, United States.