Abstract #0125

Automated motion correction of multi-slice fetal brain MRI using a deep recursive framework

Wen Shi^1,2,3, Jiwei Sun¹, Yamin Li³, Cong Sun⁴, Tianshu Zheng¹, Yi Zhang¹, Guangbin Wang⁴, and Dan Wu¹

¹Key Laboratory for Biomedical Engineering of Ministry of Education, Department of Biomedical Engineering, College of Biomedical Engineering & Instrument Science, Zhejiang University, Hangzhou, China, ²2. Department of Biomedical Engineering, Johns Hopkins University School of Medicine, Baltimore, MD, United States, ³School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China, ⁴Department of Radiology, Shandong Medical Imaging Research Institute, Cheeloo College of Medicine, Shandong University, Jinan, China

Prenatal MRI of fetal brain is vulnerable to unpredictable fetal motion and maternal movement. The conventional registration-based motion correction methods sometimes fail in excessive motion. In this work, we proposed a learning-based scheme to estimate fetal brain motion using a deep recursive framework, which replicated the iterative slice-to-volume registration and 3D volumetric reconstruction process. The network outperformed the previous learning-based methods and with good computational efficiency compared to traditional method. It also achieved high super-resolution reconstruction accuracy on simulated motion-corrupted slices, and therefore, is promising for fetal brain MRI analysis.

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

Join Here