Evaluation Of Lung Cancer Progression Based On Artificial Intelligence 3D CT Reconstruction And Discovery Of Differentially Expressed Genes Related To Progression

Objectives:Lung cancer is the most common cause of cancer death in the world,with an estimated 1.6 million deaths every year.About 85% of patients with the same histological subtype are classified as non-small cell lung cancer,of which lung adenocarcinoma and lung squamous cell carcinoma are the most common subtypes.Lung cancer is a molecularly heterogeneous disease,and understanding its biology is essential for researching effective treatments.The treatment of lung cancer has changed from traditional cytotoxic drug chemotherapy to personalized precision medicine,According to the tumor driver genes mutations of patients with lung cancer and the expression of programmed death receptor 1(PD-1)or programmed death ligand 1(PD-L1),the corresponding benefit can be obtained from targeted therapy and immune checkpoint blockers to achieve the best effect and minimum Adverse reactions.Imaging omics is an increasingly important technical means in the clinical diagnosis and treatment process,providing important diagnosis and treatment information for the clinic.At present,imaging omics is still an important examination method for lung cancer patients to diagnose,analyze the condition,judge the prognosis of the patient,and evaluate the treatment effect.Imaging features can provide tumor density,shape,size,and texture features,which may be directly or indirectly related to clinical manifestations,laboratory test results,genomics or proteomics analysis results.Use imaging omics to assess tumor size,predict patient prognosis,find the molecular basis that affects the progress of lung cancer patients based on genetic difference analysis,and help individualized treatment of non-small cell lung cancer provide better diagnosis and treatment plans,so as to achieve precision medicine.A recent study showed that for early-stage non-small cell lung cancer,tumor volume is an independent prognostic factor of long-term diseasefree survival(DFS)and overall survival(OS),and tumor volume can better predict early-stage non-small cell lung cancer the prognosis.In this study,a combination of artificial intelligence 3D CT and Genomics was used to semi-automatically cut CT images of patients with non-small cell lung cancer without distant metastasis to calculate the tumor volume,group them according to the tumor volume,and analyze the differential genes between each group to find differential genes that affect tumor growth,explore their molecular basis that affects the progression of lung cancer patients,and further analyze potential genes that affect prognosis,providing new options for targeted therapy at the molecular basis of nonsmall cell patients.Method:A total of 80 cases of non-small cell lung cancer without distant metastasis were included in this study,download the chest CT image data of NSCLC patients before treatment in the NSCLC-Radiomics-Genomics dataset from the cancer imaging database(https://public.cancerimagingarchive.net/nbia-search/),and download the corresponding gene expression data from GEO(Gene Expression Omnibus)(https://ncbi.nlm.nih.gov/geo/query/ acc.cgi? acc= GSE58661).Use the 3Dslicer’s NVIDIA AI Assisted Annotation expansion package to semiautomatically cut the tumor on the CT images of the early stage NSCLC patients before treatment to obtain the 3D model of the tumor,and use the Model module in the 3D-slicer to obtain the volume of the cut tumor.Divided into 2 groups according to the tumor volume size of reconstructed CT.The difference in gene expression between the two groups was analyzed by GEO2 R,and the differential gene with p value <0.05(adj.P.Val)and |logFC| ≥ 2 after correction was obtained,and the screening the candidate genes come out for analysis.Result:In non-small cell lung cancer without distant metastasis,SFTPC was up-regulated in volume ≤27.938cm~3.SFTPC is a gene that has a strong correlation with tumor size(adj.P.Val=0.01953,LogFc= 2.613963)..Conclusions:In this study,artificial intelligence imaging semi-automated cutting methods are used to find the molecular basis that affects the size of nonsmall cell lung cancer tumors,and to explore the inhibitory or driving factors of tumor partial development.SFTPC may inhibit tumor growth,the restricted expression of tumor suppressor genes may be one of the key factors in promoting local tumor development,and SFTPC is expected to become a potential gene for predicting the prognosis of lung cancer patients.