Study Background
The composition of the tumor immune microenvironment (TME) is complex and challenging to quantify manually.• Machine learning (ML) algorithms can be used to characterize the spatial distribution of cells and tissue regions of the TME from digitized H&E stained whole slide images (WSI) of multiple cancer types.
• Based on ML-based TME characterization, we extracted TME-associated human interpretable features (HIFs) to generate an atlas characterizing the TME in several cancer types, including bladder cancer, breast cancer, and non-small cell lung cancer (NSCLC). We term this atlas Tumor Immune Microenvironment Atlas Project (TIMAP). Conway et al.
