Imaging Biomarkers and Gene Expression Data Correlation Framework for Lung Cancer Radiogenomics Analysis Based on Deep Learning

Precision medicine, a popular treatment strategy, has become increasingly important to the development of targeted therapy. To correlate medical imaging with prognostic and genomic data, researches in radiomics and radiogenomics have provided many pre-defined image features to describe image information quantitatively or qualitatively. However, in previous researches, there are only statistical results which prove high correlation among multi-source medical data, but those can't give intuitive and visual result. In this paper, a deep learning based radiogenomic framework is provided to construct the linkage from lung tumor images to genomic data and implement generation process in turn, which form a bi-direction framework to map multi-source medical data. The imaging features are extracted from autoencoder under the condition of genomic data. It can obtain much more relevant features than traditional radiogenomic methods. Finally, we use generative adversarial network to transform genomic data onto tumor images, which gives a cogent result to explain the linkage between them. As a result, our framework provides a deep learning method to do radiogenomic researches more functionally and intuitively.

Automated summary

This study presents a deep learning-based radiogenomic framework that can provide more relevant features and vivid results to intuitively demonstrate the connections among medical data.