SELDI-TOF-MS And ANN To Eestablish Diagnostic Pattern Of Lung Cancer

Objective:Application of surface-enhanced laser desorption ionization time of flight mass spectrometry(SELDI-TOF-MS) analysis of primary lung cancer group(hereinafter referred to as lung cancer) and control group populations(including healthy controls and benign pulmonary diseases group) of serum protein fingerprinting changes in serum of patients with primary lung cancer screening of differentially expressed proteins.The use of artificial neural network model of molecular diagnostic SELDI protein construct can be used for early diagnosis of primary lung cancer A new method of lung cancer in order to facilitate early diagnosis and treatment of patients with lung cancer improve the 5-year survival rate..Methods:The study include 102 sera(54 case of lung cancer,13case of benign pulmonary diseases,and 35 case of normal controls ) used for identification of statistically significant peaks as well as for ANN model development and be used for test of the ANN model to validate its diagnostic efficiency and clinical value.The SELDI-TOF-MS and golden protein chip were performed to detect mass spectrogram for serum protein signature analysis.Then,The different expressed markers were screened from the maps by Biomarker Wizard 3.1 software and further to build ANN model.ROC curve was used to evaluate its diagnostic value.Result:Total of 106 different expressed protein peaks were detected between the group of lung cancer and normal contrast group.Five masses with an average mass of 2955,5948 were up-regulated in lung cancer compared with normal contrast group,and mass of 4651,9257 were down-regulated.Five specific protein((M/Z 2955,3284,4651,5948,9257 )were chosen to develop the artificial neural network diagnostic model.The model was tested and yielded a sensitivity of 90.7%,a specificity of 91.4%,a positive predictive value 94.2%,a negative predictive of 86.4%,a accuracy of 91%).Total of 4 different expressed protein peaks were detected between the group of lung cancer and benign pulmonary diseases group.The four specific protein((M/Z 3371,2908,2831,2756 )were chosen to develop the artificial neural network diagnostic model.The model was tested and yielded a sensitivity of 89.1%,a specificity of 58.3%,a positive predictive value 90.7%,a negative predictive of 53.8%,a accuracy of 83.6%.Conclusion:The technique of SELDI-TOF-MS is high throughput research method in proteomics with superiorities of automation,rapidness paucity of samples sensitivity and specificity.This study selected the artificial neural network model of protein molecular diagnosticâ… (M/Z 2955,3284,4651,5948,9257)be able to accurately distinguish between lung cancer and non-lung cancer.Selected artificial neural network model of molecular diagnostic proteinâ…¡(M/Z3371,2908,2831,2756) be able to distinguish between benign pulmonary diseases and lung cancer.SELDI-TOF-MS technology in the diagnosis of lung tumor-specific proteins and serum screening markers have important clinical value,can be widely used in clinical.