Meeting Banner
Abstract #4933

Quantitative Susceptibility Mapping using a Deep Learning prior

Zhe Liu1,2, Jinwei Zhang1,2, Shun Zhang2, Pascal Spincemaille2, Thanh Nguyen2, and Yi Wang1,2

1Cornell University, Ithaca, NY, United States, 2Weill Cornell Medical College, New York, NY, United States

A Bayesian method is proposed by formulating deep learning outcome as a regularization in QSM reconstruction. It enforces the fidelity between the network generated QSM and the measured inhomogeneity field. Preliminary results indicate both quantitative and qualitative improvement over QSM by deep learning alone.

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

Join Here