Multiple genetic mutations are associated with the outcomes of patients with non-small cell lung cancer (NSCLC) after using tyrosine kinase inhibitor, but the cost for detecting multiple genetic mutations is high. Few studies have investigated whether multiple genetic mutations can be simultaneously detected based on image features in patients with NSCLC. We developed a machine learning-derived radiomics approach that can simultaneously discriminate the presence of EGFR, KRAS, ERBB2, and TP53 mutations on CT images in patients with NSCLC. These findings suggest that machine learning-derived radiomics may become a noninvasive and low-cost method to screen for multiple genetic mutations in patients with NSCLC before using next-generation sequencing tests, which can help to improve individualized targeted therapies. READ ARTICLE
Cancers DOI:10.3390/cancers13081814
Authors: Tiening Zhang, Zhihan Xu, Guixue Liu, Beibei Jiang, Geertruida H. de Bock, Harry J. M. Groen, Rozemarijn Vliegenthart, Xueqian Xie
