Meeting Banner
Abstract #2715

Automatic segmentation of deep grey matter structures for iron quantification

Ying Wang1, Yongsheng Chen2,3, David T Utriainen2,4, and Ewart Mark Haacke1,2,3,4

1Biomedical Engineering, Wayne State University, Detroit, MI, United States, 2The MRI Institute for Biomedical Research, Bingham Farms, MI, United States, 3Radiology, Wayne State University, Detroit, MI, United States, 4Magnetic Resonance Innovations Inc., Bingham Farms, MI, United States

Quantitative susceptibility mapping (QSM) is a promising iron quantification method for assessing subcortical deep gray matter (SGM) in various neurodegenerative diseases. The accuracy of the measurement depends largely on the accuracy of the structural segmentation. Manually drawn regions-of-interest from a well-trained specialist are often the best but are very time-consuming. In this work, we propose an automatic segmentation method for DGM iron quantification by taking advantage of a hybrid image approach combining T1W images and QSM data. Preliminary results on 5 stroke patients presented an overall 77.8±5.8% Dice coefficient compared to the manually drawn ground truth. The measured susceptibility of the DGM showed good agreement between both methods.

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

Join Here