Computer Methods and Programs for Medical Decision Making

In this book, a strategy for medical decision diagnosis in cardiac imaging with MRI was depicted. This approach stipulates to conceive a 5D model, which depends on five dimensions: anatomical structure of the heart in 3D, temporal dimensions as well as a functional dimension of blood flow for the detection of stenosis and valvular regurgitation. Furthermore, a comparative study of the practical tools, which exploit the fifth dimension of flow for the deduction of the medical inference in the clinical routine. This contribution was considered for a stenosis aorta, which consists to make a 3D model and to solve the Navier-stokes equations for the laminar and viscous blood fluid to deduce the proposed model 5D (3D+time+flow). The extraction of measurements (field of the vortex, masses of flow, static pressure, Reynolds number) based on the fifth dimension is used to classify the area and the degree of stenosis. With the increasing demand for high-resolution simulations, it has become important to study the cost and response time of digital solvers that could take benefit of recent architectures including multi-core processors. In this work, the simulation, an efficient solver for the resolution based on the heterogeneous (CPU/GPU) architectures of Navier-Stokes (NS) equations relating to flows of incompressible fluids was assessed.