Meeting Banner
Abstract #0090

Deep learning based acceleration of multi-contrast MRI examinations by acquiring contrast and sharing inter-contrast structure information

Sudhanya Chatterjee1, Suresh Emmanuel Joel1, Sajith Rajamani1, Shaik Ahmed1, Uday Patil1, Ramesh Venkatesan1, and Dattesh Dayanand Shanbhag1
1GE Healthcare, Bangalore, India

We propose a method to accelerate multi-contrast MRI examination for a subject. We accelerate typically longer scans of MR exam such as FLAIR-T1, FLAIR-T2 by only acquiring its contrast information and sharing structure information from a reference scan from the same exam which is fully sampled (typically with the lowest acquisition time e.g. T2FSE). The resulting view-shared images have systematic artifacts which are then removed by a deep learning module trained on more than 200 cases. We observed high quality reconstruction (SSIM>0.9) for both healthy control and pathology cases with acceleration factors of 2x and 3x for FLAIR-T1 and FLAIR-T2.

This abstract and the presentation materials are available to members only; a login is required.

Join Here