Pulmonary imaging omics research

Problem to be solved: 1. Manual ROI delineation takes time, and semi-automatic delineation accuracy is poor. A universal ROI delineation software with high semi-automatic delineation accuracy is needed. AK software runs slightly slower, and large-scale data processing needs to be accelerated. 3. The result presentation lacks standardization. 4. The software of various companies lacks standardization. The planned goal is to establish a comprehensive imaging omics analysis software for pulmonary nodules and lung cancer based on a large amount of data on pulmonary nodules and lung cancer cases in our department and previous research achievements, and apply it to daily clinical work. Using imaging omics to distinguish between benign and malignant solitary pulmonary nodules, and analyzing and comparing the imaging omics characteristics of different pathological types of lung cancer and lung cancer with different degrees of differentiation. Using imaging omics to predict and use imaging omics to distinguish between benign and malignant solitary pulmonary nodules, And analyze and compare the imaging omics characteristics of different pathological types of lung cancer and different degrees of differentiation of lung cancer to determine whether there is a difference in the characteristics between non-small cell lung cancer and small cell lung cancer. Use imaging omics to predict the EGFR mutation status of lung cancer (approximately 300 cases in this study), thus providing more valuable information for clinical treatment.