University of Wisconsin–Madison

Deep Learning: Recent Applications in Medical Imaging

Picture of Fang Liu, PhD

Speaker: Fang Liu (UW)

Seminar Title: Deep Learning: Recent Applications in Medical Imaging

Date: March 23

Time: 4pm

Location: ME1106

Summary: This talk will present an overview of Deep Learning (DL) and discuss some recent successful applications in medical imaging. One aim is to draw connections between DL methods such as convolutional neural network (CNN), convolutional encoder-decoder (CED), cycle-consistent adversarial neural network (Cycle-GAN) and medical applications including image reconstruction, multi-modality image synthesis, image segmentation and computer-assisted image diagnosis. Dr. Liu will present some of his recent work using DL for medical imaging applications and will discuss relevant DL methods and their strengths and limitations. The talk will conclude with a discussion of open problems in DL that are particularly relevant in medical imaging and the potential challenges of DL in this emerging field.

Speaker Bio: Fang Liu’s research interests focus on magnetic resonance (MR) imaging, more specifically for MR image reconstruction, pulse sequence design, quantitative mapping, and MR image analysis. Dr. Liu is currently an assistant scientist in the Department of Radiology, UW-Madison with particular focus on deep learning and medical imaging. Dr. Liu’s projects include three main research fields. Firstly, he leads a research team to investigate advanced methods for performing fast and accurate quantitative mapping techniques for assessing tissue properties in musculoskeletal imaging. These techniques include ultra-short echo time imaging, diffusion imaging, and multi-component relaxometry. Secondly, he designs accelerated MR imaging techniques using deep learning, compressed sensing and MR fingerprinting. Finally, he investigates novel deep learning techniques for medical image applications and recently successfully translated deep learning methods into multiple research projects for automated image segmentation, PET/MR attenuation correction and MR-only radiation therapy.