Miranda Lab Machine Learning & Image-based profiling

Projects

Cross-Modal Generative AI for Spatial Transcriptomics
Cross-Modal Generative AI for Spatial Transcriptomics Uncertainty-aware modeling across tissue morphology and gene expression

The project is developing and evaluating uncertainty-aware generative models linking tissue morphology and spatial gene expression while preserving biological heterogeneity.

Self-Supervised Foundation Models for Skin Barrier Assessment
Self-Supervised Foundation Models for Skin Barrier Assessment Stratum corneum nanotexture as a scalable digital biomarker

This project develops self-supervised foundation models from nearly 25,000 atomic force microscopy images to investigate stratum corneum nanotexture (SCN) as a non-invasive digital biomarker across dermatological conditions, ultraviolet radiation exposure, and population variation.

More

PALS - SciLifeLab KAW
PALS - SciLifeLab/KAW Linking Omics and Images with Deep Learning to Decode Cellular Responses