Objective To explore the value of digital X-radiography(DR)in the differential diagnosis of benign and malignant isolated pulmonary nodules.Methods A retrospective analysis included 80 patients with isolated pulmonary nodules from December 2020 to December 2023 in our hospital,all of whom were diagnosed with DR.The random number method was used,including physician assessment(physician evaluating imaging results)and artificial intelligence(AI)assessment(imaging results),and the value of DR in the differential diagnosis of benign and malignant nodules was analyzed based on the pathological diagnosis results.Results Among the 80 patients,the nodule diameter was 0.9～2.7cm,mean(1.7±0.3)cm.26 cases nodules were in upper right lobe,13 cases nodules were in right middle lobe,10 cases nodules were in right lower lobe,19 nodules were in left upper lobe,12 nodules were in left lower lobe.43 cases had benign nodules,37 cases had malignant nodules,including 32 cases with adenocarcinoma and 6 cases with squamous carcinoma.The detection rate of artificial intelligence(90.00%)was significantly higher than that of conventional diagnosis(78%),and the difference was statistically significant(P<0.05).The AI sensitivity,specificity,accuracy,positive predictive value,and negative predictive value were all higher than HR-T2WI,IVIM-D,and conventional diagnosis.Routine diagnosis Kappa value was=0.58 and AI diagnosis Kappa value was=0.92.The calcification rate,exfoliation rate,and burr sign rate of patients with malignant nodules were higher than those of benign nodules,and the difference was significant(P<0.05).Conclusion DR can differentially diagnose benign and malignant isolated pulmonary nodules,and artificial intelligence evaluation can significantly improve diagnostic efficiency.
关键词
数字X射线摄影/良恶性结节/孤立性肺结节/人工智能
Key words
Digital X-ray photography/Benign and malignant nodules/Isolated pulmonary nodules/Artificial intelligence
维普
万方数据