Support on developing AI training pipeline for sparse sample 4D flow reconstruction

Ongoing:	Yes
Start:	2024-02-14

4D flow is a medical imaging technique that captures blood flow through the body, such as in the heart and aorta, over time. This enables accurate modeling of blood velocity and pressure, making it an invaluable tool for diagnosing various cardiac conditions.

However, the imaging process is time-consuming, which can be challenging for patients. This project aims to accelerate the time of imaging by reducing the number of samples which the scanner takes, leading to a sparse sampling. AI techniques are then used to reconstruct the full image data with acceleration factors up to 20x which have the potential to vastly reduce the time a patient has to be subjected to the examination.

AIDA Data Hub supports this project by assisting with training pipeline development.