Abstract #4714

SSTN: Self-supervised Triple-Network with Multi-scale Dilated-Convolution for Fast MRI Reconstruction

Yuekai Sun¹, Jun Lv¹, Weibo Chen², He Wang^3,4, and Chengyan Wang⁴

¹School of Computer and Control Engineering, Yantai University, Yantai, China, ²Philips Healthcare, Shanghai, China, ³Institute of Science and Technology for Brain-inspired Intelligence, Fudan University, Shanghai, China, ⁴Human Phenome Institute, Fudan University, Shanghai, China

Synopsis

To solve the problem that fully-sampled reference data is difficult to acquire, we proposed a self-supervised triple-network (SSTN) for fast MRI reconstruction. Each pipeline of SSTN is composed of multiple parallel ISTA-Net blocks which consists of different scales dilated convolution layers. The results demonstrated that the proposed SSTN can generate better quality reconstructions than competing methods at high acceleration rates.

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

Join Here