Zikang Xu (徐梓康)

I am a fourth-year Ph.D candidate at School of Biomedical Engineering, University of Science and Technology of China, Suzhou, China, majoring in Medical Image Analysis, Fairness, Trustworthy AI.

Currently, I am doing research about developing fair algorithms for medical image analysis supervised by Prof. S. Kevin Zhou (周少华) at the MIRACLE Center.

Before I came to USTC, I received my bachelor degree and master degree (Supervisor: Prof. Zhongze Gu (顾忠泽)) from Southeast University in 2018 and 2021, respectively.

I am seeking for post-doctoral positions in the field of medical image analysis. If you have interest in my research, please feel free to contact me.

Email  /  Google Scholar  /  GitHub  /  CV

profile photo

Selected Publications
FairAdaBN: Mitigating unfairness with adaptive batch normalization and its application to dermatological disease classification
Zikang Xu, Shang Zhao, Quan Quan, Qingsong Yao, S. Kevin Zhou
International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), 2023.
  code /arXiv /bibtex /poster

A novel FairAdaBN module which process samples in different sensitive attributes differently, to ensure model fairness in two dermatological datasets.

Addressing Fairness Issues in Deep Learning-Based Medical Image Analysis: A Systematic Review
Zikang Xu, Jun Li, Qingsong Yao, Han Li, Mingyue Zhao, and S. Kevin Zhou
npj Dig. Med., 2024.
  code /Springer Nature /bibtex

A survey of fairness in medical image analysis, including concepts, algorithms, evaluations, and challenges.

FairMedFM: Fairness Benchmarking for Medical Imaging Foundation Models
Ruinan Jin*, Zikang Xu*, Yuan Zhong*, Qingsong Yao, Qi Dou†, S. Kevin Zhou†, Xiaoxiao Li†
Annual Conference on Neural Information Processing Systems (NeurIPS) Dataset & Benchmark Track, 2024.
  code /arXiv /bibtex

An evaluation of fairness on medical foundation models, involving 20 foundation models on 17 medical datasets.

APPLE: Adversarial Privacy-aware Perturbations on Latent Embedding for Unfairness Mitigation
Zikang Xu, Fenghe Tang, Quan Quan, Qingsong Yao, S. Kevin Zhou
arxiv, 2023.
  code /arXiv /bibtex

Mitigating unfairness in deployed U-Net-like and SAM-family segmentation models by latent perturbation.

Inspecting Model Fairness in Ultrasound Segmentation Tasks
Zikang Xu, Fenghe Tang, Quan Quan, Jianrui Ding, Chunping Ning, S. Kevin Zhou
arxiv., 2023.
arXiv /bibtex

First attempt in evaluating fairness issues in ultrasound image segmentation.

Slide-SAM: Medical SAM Meets Sliding Window
Quan Quan, Fenghe Tang, Zikang Xu, Heqin Zhu, S. Kevin Zhou
Medical Imaging with Deep Learning (MIDL)., 2023, oral.
  code /arXiv /bibtex

Modify Segment Anything Model (SAM) to 3D medical image segmentation.

OFELIA: Optical Flow-based Electrode LocalIzAtion
Xinyi Wang, Zikang Xu, Qingsong Yao, Yiyong Sun, S. Kevin Zhou
Medical Imaging with Deep Learning (MIDL)., 2023.
OpenReview /bibtex

Optical-flow-based electrode localization in catheter ablation.

SIX-Net: Spatial-Context Information miX-up for Electrode Landmark Detection
Xinyi Wang, Zikang Xu, Heqin Zhu, Qingsong Yao, Yiyong Sun, S. Kevin Zhou
International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), 2024, early accept.
Springer /bibtex

Electrode localization in catheter ablation by spatial-context information mixup.

LACOSTE: Exploiting stereo and temporal contexts for surgical instrument segmentation
Qiyuan Wang, Shang Zhao, Zikang Xu, S. Kevin Zhou
Medical Image Analysis, Volume 99, 1 January 2025, 103387.
ScienceDirect /bibtex

LACOSTE exploits stereo and temporal contexts for surgical instrument segmentation by applying three modules and alignment loss to make improvement from different views.

A storm in a teacup--A biomimetic lung microphysiological system in conjunction with a deep-learning algorithm to monitor lung pathological and inflammatory reactions
Zaozao Chen, Jie Huang, Jing Zhang, Zikang Xu, Qiwei Li, Jun Ouyang, Yuchuan Yan, Shiqi Sun, Huan Ye, Fei Wang, Jianfeng Zhu, Zhangyan Wang, Jie Chao, Yuepu Pu, Zhongze Gu
Biosensors and Bioelectronics, Volume 219, 1 January 2023, 114772.
ScienceDirect /bibtex

A deep-learning algorithm was developed to characterize the activation of cells in Lung-MPS, which could provide an improved and more biomimetic sensory system for the study of COVID-19 and other high-risk infectious lung diseases.


Research Experience
  • [2023/06 - 2023/10] Research Intern, Fuwai Hospital, CAMS & PUMC

Professional Services
  • Journal Reviewer: IEEE TMI
  • Conference Reviewer: MICCAI 2024, BIBM 2024


Modified from Heqin Zhu