Quantitative CT analysis of pulmonary nodules for lung adenocarcinoma risk classification based on an exponential weighted grey scale angular density distribution feature

•A greatly robustness and highly effective pulmonary nodule classification system.•The reference map is constructed using an integral image and labelled the map by K-means. Then, the grey density distribution feature is generated.•Furthermore, we proposed the feature extraction method for pulmonary nodule image by designing an exponential weighted multi-angular histogram to describe each component of the grey density distribution features.•The proposed feature combined with Random Forest model to classify lung nodule to four categories: Atypical Adenomatous Hyperplasia (AAH), Adenocarcinoma In Situ (AIS), Minimally Invasive Adenocarcinoma (MIA), and Invasive Adenocarcinoma (IA).