Abstract #2724

Data-driven regularized inversion (DRI) for improved QSM+qBOLD based CMRO2 Mapping: a feasibility study in healthy subjects and ischemic stroke patients

Junghun Cho¹, Shun Zhang², Youngwook Kee³, Pascal Spincemaille³, Thanh Nguyen³, Simon Hubertus⁴, Ajay Gupta³, and Yi Wang^1,3

¹Biomedical Engineering, Cornell University, Ithaca, NY, United States, ²Radiology, Tongji Hospital, Wuhan, China, ³Radiology, Weill Cornell Medical College, New York, NY, United States, ⁴Computer Assisted Clinical Medicine, Heidelberg University, Mannheim, Germany

We propose the use of machine-learning to improve the accuracy of a QSM+qBOLD model based Cerebral metabolic rate of oxygen (CMRO2) and oxygen extraction fraction (OEF) mapping. The proposed method, data-driven regularized inversion or DRI, significantly outperformed, in simulation, the current method at all SNR levels. In n=11 healthy subjects, uniform OEF maps were obtained as expected. In n=18 ischemic stroke patients, low OEF regions were clearly located within the lesion region as defined by DWI.

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

Join Here