Genotype-Guided Radiomics Signatures for Recurrence Prediction of Non-Small Cell Lung Cancer

Non-small cell lung cancer (NSCLC) is a serious disease and has high recurrence rate after surgery. Recently, many machine learning methods have been proposed for prediction. The using gene expression data achieve accuracy rates but expensive. While, the radiomics features computer tomography (CT) image cost-effective method, their not competitive. In this paper, we propose genotype-guided method (GGR) obtaining prediction at low cost. We used public radiogenomics dataset of NSCLC, which includes CT images data. Our two steps that uses models. first model estimation model, to estimate from deep extracted images. second predict estimated gene. GGR designed based on hybrid fusion handcrafted- learning-based features. experiments demonstrated can be improved significantly 78.61% (existing method) 79.09% (ResNet50) 83.28% by GGR.