基于深度学习校正头颈部动脉和冠状动脉MR血管成像研究进展  

Research progresses of deep learning for correcting MR angiography of head and neck arteries and coronary arteries

作　　者：曹波 卢洁[1,2] CAO Bo;LU Jie(Department of Radiology and Nuclear Medicine,Xuanwu Hospital,Capital Medical University,Beijing 100053,China;Beijing Key Laboratory of Magnetic Resonance Imaging Brain and Brain Informatics,Beijing 100053,China)

机构地区：[1]首都医科大学宣武医院放射与核医学科,北京100053 [2]磁共振成像脑信息学北京市重点实验室,北京100053

出　　处：《中国医学影像技术》2024年第10期1599-1602,共4页Chinese Journal of Medical Imaging Technology

基　　金：国家自然科学基金(82130058、81974261)。

摘　　要：MR血管成像(MRA)是非侵入性影像学技术,具有无辐射和高分辨率等优点,对评估血管狭窄等具有重要作用;但血管搏动和呼吸运动等可导致运动伪影,从而降低图像质量。随着深度学习(DL)技术的发展,基于此的图像校正方法逐渐广泛用于图像后处理。本文就DL用于校正头颈部动脉和冠状动脉MRA研究进展进行综述。MR angiography(MRA)is a non-invasive imaging technique with advantages such as non-radiation and high resolution,which plays an important role in evaluating vascular stenosis.However,vascular pulsation and respiratory motion can lead to motion artifacts,thereby reducing image quality.With the development of deep learning(DL)technology,image correction based on this has gradually become widely used in image post-processing.The research progresses of DL for correcting MRA of head and neck arteries and coronary arteries were reviewed in this article.

关 键 词：动脉 深度学习 磁共振血管造影术 

分 类 号：R654.3[医药卫生—外科学] R445.2[医药卫生—临床医学]