Contact information
Kexin Xu
DPhil Student in Medical Science
- MRC WIMM Centre of Computational Biology
I am a biologist who taught myself machine learning and deep learning. I am interested in using AI tools to interrogate biological data to derive insights that aid diagnosis and prognosis. I have developed/used AI workflows for natural language processing, multi-omics analysis, epidemiology, and medical imaging (X-ray & histology images).
In my PhD, I am developing a deep learning workflow to predict spatial transcriptomics from readily available histology images to bypass the expensive and time-consuming experimental protocol. I am interested in using AI to bridge multi-omics and medical imaging to improve clinical workflow.
Key publications
Spatial fibroblast niches define Crohn's fistulae.
Journal article
McGregor C. et al, (2026), Nature, 649, 703 - 712
ssessing Metabolic Ageing via DNA Methylation Surrogate Markers: A Multicohort Study in Britain, Ireland and the USA.
Journal article
Xu K. et al, (2025), Aging Cell, 24
Designing a computer-assisted diagnosis system for cardiomegaly detection and radiology report generation.
Journal article
Zhu T. et al, (2025), PLOS Digit Health, 4
LAVASET: Latent Variable Stochastic Ensemble of Trees. An ensemble method for correlated datasets with spatial, spectral, and temporal dependencies.
Journal article
Kasapi M. et al, (2024), Bioinformatics, 40
Recent publications
Spatial fibroblast niches define Crohn's fistulae.
Journal article
McGregor C. et al, (2026), Nature, 649, 703 - 712
Correction: Spatial fibroblast niches define Crohn's fistulae.
Journal article
McGregor C. et al, (2025), Nature, 648
ssessing Metabolic Ageing via DNA Methylation Surrogate Markers: A Multicohort Study in Britain, Ireland and the USA.
Journal article
Xu K. et al, (2025), Aging Cell, 24
Designing a computer-assisted diagnosis system for cardiomegaly detection and radiology report generation.
Journal article
Zhu T. et al, (2025), PLOS Digit Health, 4
LAVASET: Latent Variable Stochastic Ensemble of Trees. An ensemble method for correlated datasets with spatial, spectral, and temporal dependencies.
Journal article
Kasapi M. et al, (2024), Bioinformatics, 40
