Researchers have developed BigCell, a novel approach capable of generating gigapixel-resolution whole-slide images (WSIs) of histopathology tissue. Unlike previous models that operated on smaller image patches, BigCell generates coherent images at native gigapixel scale by training a flow-matching transformer on over 120,000 WSIs. The system can produce images up to 32,768x32,768 pixels and incorporates an adaptive scheduling strategy to accelerate generation. Additionally, BigCell introduces text-to-image synthesis at 8k resolution, directly from pathology text reports, which has been shown to improve slide-level classification performance. AI
IMPACT Enables generation of high-resolution medical images from text, potentially improving diagnostic accuracy and training data availability.
RANK_REASON The cluster describes a new research paper detailing a novel AI model for generating high-resolution medical images. [lever_c_demoted from research: ic=1 ai=1.0]
Read on Hugging Face Daily Papers →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →