Microsoft researchers have unveiled GigaTIME, a powerful artificial intelligence system designed to model the complex environment surrounding tumors.

Instead of relying solely on expensive and sophisticated laboratory tests, the system can analyze standard pathology slides and generate detailed molecular insights into what is happening inside tumors.

The AI was trained with data from approximately 40 million cells, connecting tissue images with information on protein levels. It was then applied to data from over 14,000 cancer patients, generating hundreds of thousands of virtual molecular images across 24 different cancer types.

This enormous “virtual population” allowed researchers to discover over 1,200 new biological associations, linking proteins, biomarkers, tumor progression, and patient survival.

What makes this especially exciting is that studying the tumor microenvironment typically requires extremely expensive and slow imaging techniques. GigaTIME can simulate similar insights using AI, allowing scientists to analyze cancer biology on a population scale, something that was previously virtually impossible.

The bigger picture: artificial intelligence is beginning to act as a discovery engine for biology.

All cancers may soon be curable.