MSU-Net: Multiscale Statistical U-Net for Real-time 3D Cardiac MRI Video Segmentation

Abstract

Cardiac magnetic resonance imaging (MRI) is an essential tool for MRI-guided surgery and real-time intervention. The MRI videos are expected to be segmented on-the-fly in real practice. However, existing segmentation methods would suffer from drastic accuracy loss when modified for speedup. In this work, we propose Multiscale Statistical U-Net (MSU-Net) for real-time 3D MRI video segmentation in cardiac surgical guidance. Our idea is to model the input samples as multiscale canonical form distributions for speedup, while the spatio-temporal correlation is still fully utilized. A parallel statistical U-Net is then designed to efficiently process these distributions. The fast data sampling and efficient parallel structure of MSU-Net endorse the fast and accurate inference. Compared with vanilla U-Net and a modified state-of-the-art method GridNet, our method achieves up to 268% and 237% speedup with 1.6% and 3.6% increased Dice scores.

arXiv commentMICCAI19

References in corpus (1)

• Accelerated Computing in Magnetic Resonance Imaging -- Real-Time Imaging Using Non-Linear Inverse Reconstruction

Cited by in corpus (7)

• U-Net and its variants for medical image segmentation: theory and applications
• Modality specific U-Net variants for biomedical image segmentation: A survey
• Positional Contrastive Learning for Volumetric Medical Image Segmentation
• DLBC: A Deep Learning-Based Consensus in Blockchains for Deep Learning Services
• Contrastive Learning with Temporal Correlated Medical Images: A Case Study using Lung Segmentation in Chest X-Rays
• Towards Cardiac Intervention Assistance: Hardware-aware Neural Architecture Exploration for Real-Time 3D Cardiac Cine MRI Segmentation
• EchoCP: An Echocardiography Dataset in Contrast Transthoracic Echocardiography for Patent Foramen Ovale Diagnosis