Meeting Banner
Abstract #3932

Estimating Cerebral Blood Flow from BOLD Signal Using Deep Dilated Wide Activation Networks

Danfeng Xie1, Danfeng Xie1, Yiran Li1, Hanlu Yang1, Li Bai1, Donghui Song 2, Yuanqi Shang2, Qiu Ge2, and Ze Wang3
1Temple University, Philadelphia, PA, United States, 2Hangzhou Normal University, Hangzhou, China, 3University of Maryland School of Medicine, Philadelphia, MD, United States

The purpose of this study was to synthesize Arterial Spin Labeling (ASL) cerebral blood flow (CBF) signal from blood-oxygen-level-dependent (BOLD) fMRI signal using deep machine learning (DL). Experimental results in the dual-echo Arterial Spin Labeling sequence show that the BOLD-to-ASL synthesize networks, the BOA-Net will yield similar cerebral blood flow value to that measured by ASL MRI and the cerebral blood flow maps produced by BOA-Net will show higher Signal-to-Noise Ratio (SNR) than that from ASL MRI.

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

Join Here