Yukun Zhou (周玉昆)

I am a research fellow at UCL, working with Daniel C. Alexander and Pearse A. Keane. I work on generalisable medical image computing and large-scale translational research #opensourse.

[scholar][Github]

Main Research Focus

Selected publications

  1. CF-Loss: Clinically-relevant feature optimised loss function for retinal multi-class vessel segmentation and vascular feature measurement. Medical Image Analysis 2024. [paper][code]

  2. A foundation model for generalisable disease detection from retinal images. Nature 2023. [paper][code]

  3. AutoMorph: Automated Retinal Vascular Morphology Quantification via a Deep Learning Pipeline. TVST 2022. [paper][code]

  4. Learning to Address Intra-segment Misclassification in Retinal Imaging. MICCAI 2021. [paper][code]

  5. A refined equilibrium generative adversarial network for retinal vessel segmentation. Neurocomputing 2021. [paper][code]

  6. CD-Loss: Clinically driven loss function for retinal multi-class vessel segmentation and vascular feature measurement. under review.

Collaboration works

  1. Expectation maximization pseudo labels. Medical Image Analysis 2024. [paper]

  2. Retinal optical coherence tomography features associated with incident and prevalent Parkinson disease. Neurology 2023. [paper]

  3. MisMatch: Calibrated Segmentation via Consistency on Differential Morphological Feature Perturbations with Limited Labels. IEEE TMI 2023. [paper]

  4. Association Between Retinal Features From Multimodal Imaging and Schizophrenia. JAMA psychiatry 2023. [paper]

  5. Bayesian Pseudo Labels: Expectation Maximization for Robust and Efficient Semi-Supervised Segmentation. MICCAI 2022 best paper runner up. [paper]

  6. Prognostic Imaging Biomarker Discovery in Survival Analysis for Idiopathic Pulmonary Fibrosis. MICCAI 2022. [paper]

  7. Progressive Subsampling for Oversampled Data - Application to Quantitative MRI. MICCAI 2022. [paper]

  8. Learning Morphological Feature Perturbations for Calibrated Semi-Supervised Segmentation. MIDL 2022. [paper]

Contacts

ykzhoua@gmail.com and yukun.zhou.19@ucl.ac.uk