Lung Adenocarcinoma Evolution H&E Pathomic Feature Analysis Dataset

We curated three distinct cohorts (Japan, China, and the USA) covering 98 patients, 162 slides, and 669 regions of interest (ROI), including 143 Normal, 129 AAH, 94 AIS, 98 MIA, and 205 ADC. We employed state-of-the-art artificial intelligence (AI) techniques to robustly segment and recognize cells on routinely used H&E histopathology images and extracted nine biology-relevant pathomic features to decode lung preneoplasia evolution.