iiinova — where your digital self keeps evolving

Minxing Pang

@minxingpang.bsky.social

Applied Math PhD student @upenn.edu

Minxing Pang @minxingpang.bsky.social · 15/12/2025

3/3 By leveraging generative AI, our framework significantly improves downstream tasks compared to traditional methods, including better cell type annotation and patient survival prediction. 📉✅ Authors: Minxing Pang, Tarun Kanti Roy, Xiaodong Wu, and Kai Tan. #GenerativeAI #SpatialOmics #Pathology

000

Minxing Pang @minxingpang.bsky.social · 15/12/2025

2/3 To power this, we created PRISM-12M: a massive dataset of 12M+ patches across 14 tissue types. 📂 TissueCraftAI uses this to generate multi-modal spatial transcriptomics & proteomics maps from simple natural language prompts, covering both human and mouse tissues. 🐭🧍‍♂️

000

Minxing Pang @minxingpang.bsky.social · 15/12/2025

1/3 Excited to share our latest preprint: TissueCraftAI! 🚀 It's the FIRST 1B-parameter spatial omics model built on ControlNet & Stable Diffusion. 🧬🤖 Moving beyond CNNs, it enables query-driven prediction of molecular maps directly from histology. 🔗 www.biorxiv.org/content/10.6...

biorxiv.org

Query-driven generative AI synthesizes multi-modal spatial omics from histology

Spatial omics technologies offer unprecedented insights into the cellular organization of tissues; however, they are not yet scalable for routine clinical use. In contrast, images of histological stai...

021