Haowei Xiang

I am now doing post-doctorate research at University of Michigan, ECE department and I was lucky to be co-advised by Prof. Jeffrey Fessler and Prof. Douglas Noll. My main research interests include signal and image processing, computational imaging, inverse problems, and machine learning, with a focus on the MRI.
I have been working on developing algorithms for novel silent MRI techniques, including model-based and learning-based image reconstruction, k-space sampling optimization, dynamic image reconstruction, RF optimization, and pulse sequence design. In my past project, I used deep Convolutional Neural Networks (CNN) to predict SPECT/CT scatter by levering the historical and simulation data. In addition to my research, I was the graduate student instructor for graduate-level courses including Probability and Random Process, Matrix Methods for Signal Processing, Data Analysis and Machine Learning, and Medical Imaging Systems at University of Michigan.

Click here for a quick overview of my research.

News

• [May 2024] Research: I successfully defended my Ph.D. thesis titled “Advanced Image Reconstruction and Sampling Pattern Optimization in Silent MRI”.
• [Feb 2024] Research: Joint optimization of multi-echo reconstruction and quantitative map estimation in Looping Star was accepted as an Oral presentation (Top 15%) at ISMRM.
• [Jan 2024] Research: “Model-based reconstruction for looping-star MRI” (Magnetic Resonance in Medicine, 2024) was accepted. This work presents an advanced model-based reconstruction approach for Looping-Star MRI, improving image quality and accuracy.
• [Aug 2023] Patent: Model-based reconstruction for looping-star pulse sequences in MRI was submitted as a US patent application.
• [Feb 2023] Research: Spatial-temporal reconstruction using UNFOLD in looping-star silent fMRI was accepted to ISMRM.
• [Sep 2022] Invited talk: I was honored to present Model-Based Image Reconstruction in Functional MRI using Looping-Star at the 2022 Functional MRI Symposium.
• [Feb 2022] Research: Model-based Image Reconstruction in Looping-star MRI was accepted to ISMRM.
• [Dec 2020] Research: A deep neural network for fast and accurate scatter estimation in quantitative SPECT/CT under challenging scatter conditions was accepted to the European Journal of Nuclear Medicine and Molecular Imaging.
• [Oct 2019] Research: SPECT/CT scatter estimation using a deep convolutional neural network: implementation in Y-90 imaging was accepted to the IEEE Nuclear Science Symposium and Medical Imaging Conference.