Home > People > Faculty

Faculty

Hongxiao WANG

Assistant Professor

Personal Information    

Research Interest:Biomedical image analysis, Deep Learning    

Mailbox:hwang21@cnu.edu.cn    

Office: Room 321, Teaching Building II, Main Campus.      

Education  

Ph.D. 2017-2023 University of Notre Dame, USA,  

Master 2014-2017 Beijing Computational Science Research Center, China  

Bachelor 2010-2014 University of Science and Technology of China, China  

Research Projects

Geometry aware medical image analysis

Geometry has the potential to bridge the gap between human knowledge and AI models. Medical knowledge often relies on geometrical descriptions (e.g., retinal vessel tortuosity and cell clustering density). By incorporating geometry into deep learning reasoning, AI can become more accurate by leveraging doctors' knowledge. Humans are more likely to trust AI if it can present clinically meaningful evidence.

To address this, my TMI-2023 project proposed a new hierarchical graph-based pathology image classification model that emulates doctors’ diagnostic expertise. This model effectively and comprehensively utilizes visual information, cell microenvironment, and patterns of cell community gathering.

Generation-based data augmentation    

Due to variations in staining conditions and microscopic environments, deep learning models often struggle to transfer applications to pathological image data from different hospitals. In my BIBM-2020 paper, I designed a data augmentation algorithm based on style transfer, which achieves pathological image segmentation accuracy previously requiring the use of the entire dataset with only 50% of the training data. This significantly reduces the time cost of annotating medical images for doctors.     

Biomedical knowledge structuring

Gene Ontology is a collection of concepts related to genes, and these concepts form a network of a directed acyclic graph based on classification relationships. In response to this topological relationship, I propose a new method for knowledge graph embedding in TCBB-2022, significantly improving the accuracy of protein similarity prediction.     

Papers and preprints      

【1】Hongxiao Wang, Gang Huang, Zhuo Zhao, Liang Cheng, Anna Juncker-Jensen, Máté, Levente Nagy, Xin Lu, Xiangliang Zhang, Danny Z. Chen. “CCF-GNN: A unified model aggregating appearance, microenvironment, and topology for pathology image classification”, IEEE Transactions on Medical Imaging (TMI), 2023 中科院一区    

【2】Hongxiao Wang, Hao Zheng, and Danny Z. Chen. “TANGO: A GO-term embedding based method for protein semantic similarity prediction.” IEEE/ACM Transactions on Computational Biology and Bioinformatics (TCBB), 2022. 中科院三区    

【3】Hongxiao Wang, Hao Zheng, Jianxu Chen, Lin Yang, Yizhe Zhang, and Danny Z. Chen. “Unlabeled data guided semi-supervised histopathology image segmentation.” IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 2020. EI会议,CCF- B类    

【4】Hongxiao Wang, and Haiguang Liu. “Determining complex structures using docking method with single particle scattering data.” Frontiers in Molecular Biosciences, 2017. 中科院三区    

【5】Hao Zheng, Jun Han, Hongxiao Wang, Lin Yang, Zhuo Zhao, Chaoli Wang, and Danny Z. Chen. “Hierarchical self-supervised learning for medical image segmentation based on multi-domain data aggregation.” International Conference on Medical Image Computing and Computer-Assisted Intervention, (MICCAI), 2021.

【6】Peixian Liang, Jianxu Chen, Yizhe Zhang, Hongxiao Wang, Hao Zheng, Pengfei Gu, and Danny Chen. “Intracker: An integrated detector-tracker framework for cell detection and tracking.” International Symposium on Computer-Based Medical Systems (CBMS), 2020.

【7】Zhuo Zhao, Hongxiao Wang, Yizhe Zhang, Hao Zheng, Siyuan Zhang, and Danny Z. Chen. “A coarse-to-fine data generation method for 2D and 3D cell nucleus segmentation.” International Symposium on Computer-Based Medical Systems (CBMS), 2020.  

【8】Chengtao Peng, Bin Li, Ming Li, Hongxiao Wang, Zhuo Zhao, Bensheng Qiu, and Danny Z. Chen. “An irregular metal trace inpainting network for X‐ray CT metal artifact reduction.” Medical Physics, 2020  

【9】Yang Zhang, Hongxiao Wang, Daniel Zhang, Yiwen Lu, and Dong Wang. “Riskcast: Social sensing based traffic risk forecasting via inductive multi-view learning.” IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM), 2019  

【10】Yang Zhang, Hongxiao Wang, Daniel Zhang, and Dong Wang. “Deeprisk: A deep transfer learning approach to migratable traffic risk estimation in intelligent transportation using social sensing.” International Conference on Distributed Computing in Sensor Systems (DCOSS), 2019.  

Awards & Honors

Outstanding Teaching Assistant Honorable Mention University of Notre Dame 2018

  • Address:No. 105, West Third Ring Road North, Beijing 100048
  • Copyright © Institute of Interdisciplinary Sciences, Capital Normal University. All rights reserved.
  • 京ICP备05082108号-1 京公网安备:110402430068号
ACADEMY FOR MULTIDISCIPLINARY STUDIES
Official WeChat