Deep Learning-based Image and Analysis for Photoacoustic and Magnetic Resonance Images

Date: 2022/05/05 - 2022/05/05

Dissertation Title: Deep Learning-based Image and Analysis for Photoacoustic and Magnetic Resonance Images

Speaker: Fei Feng, Ph.D. candidate at UM-SJTU Joint Institute

Time: 1:30 p.m., May 5th, 2022 ( Beijing Time)

Location: via feishu

Abstract

Image processing is crucial for better utilization of biomedical images and deep learning method has shown great potential in biomedical image processing. This dissertation investigated the deep learning-based algorithms for image recovery and analysis of photoacoustic  (PA) and magnetic resonance (MR) images. First, two image recovery algorithms were proposed for PA image recovery and 3D MR image through-plane super-resolution, respectively. Second, deep learning algorithms were designed for image segmentation and landmark localization using MR images. Third, a novel semi-supervised learning algorithm was investigated for pelvic MR image segmentation. These algorithms enable improved image quality and automated image analysis to make better use of PA and MR images.

Biography

Fei Feng received his B.S. degree in 2017 from Northwestern Polytechnical University. He is currently a Ph.D. candidate at UM-SJTU Joint Institute, Shanghai Jiao Tong University. His research interest is focused on solving biomedical image recovery and analysis problems by using machine learning and deep learning techniques.