Yunpeng Zhao

Bachelor of Engineering

Beihang University

Biography

Hi there! This is Yunpeng Zhao (赵云鹏 in Chinese). I got my bachelor’s degree in Aircraft Control and Information Engineering at Beihang University (BUAA) this June and going to take a gap year to deepen my research. I have decided to pursue further studies at the doctorate level upon successful completion of my undergraduate studies, and I am looking for an available Ph.D., self-funded Master, or RA position!

During my undergraduate study, I worked with Prof. Bo Liu on machine learning for medical image analysis and graph neural networks. My research interests lie in Medical Image Analysis, Machine Learning and Computer Vision currently and I am open to explore other topics.

Welcome to check my CV for more details.

Interests

Medical Image Analysis
Machine Learning
Graph Neural Networks
Computer Vision

Education

B.Eng in Aircraft Control and Information Engineering, 2018 - 2022

Beihang University (BUAA)

Experience

Undergraduate Researcher

Beihang University

Sep 2021 – Present Beijing

Worked with Prof. Bo Liu
Project: Interpretable Spatial Temporal Graph Convolutional Network with Self-learned Graph Structure for Early MCI Detection
Achievement: Got a paper submitted to BIBM 2022 (ccf-b), which is still under review now

Summer Research Intern

Beihang University & Zhejiang Future Technology Institute

Jul 2021 – Sep 2021 Zhejiang & Beijing

Worked on a project mentored by Yuan Zhong (Ph.D. candidate supervised by Prof. Yebin Liu from Tsinghua University)
Project: Punctate Highlight Removal for Microscopic Images Based on Pix2pixHD
Achievement: Outstanding Summer Intern, Best Project Award in Summer Internship of Beihang University and Zhejiang Future Technology Institute

Featured Publications

Yunpeng Zhao, Fugen Zhou, Bin Guo, Bo Liu

November, 2022 In arXiv, submitted to ISBI 2023

Spatial Temporal Graph Convolution with Graph Structure Self-learning for Early MCI Detection

This paper proposes a spatial temporal model for EMCI classification with a novel graph structure self-learning mechanism, which can adaptively learns the optimal spatial dependencies between brain regions. Our model exploits BOLD time series as input features directly, which provides a new perspective for rsfMRI-based preclinical AD diagnosis. Results on the ADNI database show superior performance of our method compared with state-of-the-art models. Moreover, EMCI-contributory brain regions consistent with existing literature can be excavated, proving the reliable interpretability of our framework.

Recent Publications

Quickly discover relevant content by filtering publications.

Yunpeng Zhao, Fugen Zhou, Bin Guo, Bo Liu (2022). Spatial Temporal Graph Convolution with Graph Structure Self-learning for Early MCI Detection. In arXiv, submitted to ISBI 2023.

PDF Cite

Projects

Bachelor Thesis

A Weakly-Supervised Image Classification Framework Based on Graph Convolutional Networks

Punctate Highlight Removal for Microscopic Images Based on Pix2pixHD

A novel punctate highlight removal network based on pix2pixHD for microscopic images.