A million-pixel image might only have one annotation from a professional pathologist, posing a significant challenge for all AI models. Our team has designed the world＇s first weakly supervised multiple instance AI learning framework. This framework can analyze million-pixel pathology slides to predict gene expression and prognosis in colorectal cancer patients. Additionally, it has been successfully validated across patient datasets from multiple generational cohorts.

Our AI system has successfully predicted the overall survival of colorectal cancer patients and identified clear correlations with pathological slide examinations. We also published our study in the top journal of Nature Communications. With this advanced digital pathology prediction system, physicians no longer rely on guessing patient survival rates but can instead rely on AI systems to clearly explain the relationship between patient prognosis and pathological slides.

In the study, we have successfully addressed the challenges of processing megapixel images and insufficient expert annotations. This enables digital pathology, after training with deep learning models, to assist physicians to make more precise clinical decisions. Additionally, we have validated recognition of specific gene expression features and patient prognosis prediction tasks from pathological images.