Objective: To comprehensively analyze the CT features and radiomics features of silent pheochromocytoma, establish models to explore the value of radiomics in differential diagnosis. Methods: A retrospective analysis of 97 cases of pheochromocytoma confirmed by surgery and pathology was performed. They were divided into silent group and non-silent group according to the presence or absence of hypertension and the triad of headache, palpitation and hyperhidrosis. The CT image features and radiomics features of the patients were compared and analyzed. Least absolute shrinkage and selection operator(LASSO) analysis was conducted to select the optimized feature subset. The conventional CT features, radiomics features, radiomics features combined with conventional CT features models were constructed using logistic regression analysis. Receiver operating characteristic(ROC) curve was used to analyze the diagnostic value of the model. Results: There was no significant difference in gender and age between the two groups(P>0.05). Among the conventional CT features, there were significant differences between the two groups in whether the longest diameter was greater than 5 cm, whether the shape was regular, and the range of cystic degeneration(P<0.05). There were nosignificant difference between the two groups of lesions in location, presence or absence of calcification, hemorrhage and cystic shape, the average CT attenuation of arterial phase and venous phase, homogenous enhancement or not, and enhancement mode and degree(P>0.05). The area under curve(AUC), accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of the radiomics features combined with conventional CT features model were 0.922(95%CI:0.870-0.974)、0.854、0.784、0.894、0.806、0.881 respectively, which were higher than the conventional CT feature model andradiomics features model. Conclusion: Although silent adrenal pheochromocytoma has no typical clinical symptoms, the CT findings have certain characteristics, and the radiomics features combined with conventional CT features model can significantly improve the differential efficacy.
 NEUMANN H P H,YOUNG W F,ENG C.Pheochromocytoma and paraganglioma[J].N Engl J Me,2019,381(6):552-565.
 FARRUGIA F A,MARTIKOS G,TZANETIS P,et al.Pheochromocytoma,diagnosis and treatment:review of the literature[J].Endocr Regul,2017,51(3):168-181.
 GURIJT S,MACKSON N,SOMNATH G,et al.Clinically silent giant pheochromocytoma:a case report[J].J Evol Med Dent Sci,2015,4(14):2422-2427.
 CLIFTON-BLIGH R.Diagnosis of silent pheochromocytoma and paraganglioma[J].Expert Rev Endocrinol Metab,2013,8(1):47-57.
 GUPTA A,BAINS L,AGARWAL M K,et al.Giant cystic pheochromocytoma:a silent entity[J].Urol Ann,2016,8(3):384-386.
 LAMBIN P,RIOS-VELAZQUEZ E,LEIJENAAR R,et al.Radiomics:Extracting more information from medical images using advanced feature analysis[J].Eur J Cancer,2012,48(4):441-446.
 LU Y,LI P,GAN W,et al.Clinical and pathological characteristics of hypertensive and normotensive adrenal pheochromocytomas[J].Exp Clin Endocrinol Diabetes,2016,124(6):372-379.
 CROUT J R,SJOERDSMA A.Turnover and metabolism of catecholamines in patients with pheochromocytoma[J].J Clin Invest,1964,43(1):94-102.
 LUBNER M G,SMITH A D,SANDRASEGARAN K,et al.CT texture analysis:definitions,applications,biologic correlates,and challenges[J].Radiographics,2017,37(5):1483-1503.
 THIBAULT G,FERTIL B,NAVARRO C,et al.Texture indexes and gray level size zone matrix application to cell nuclei classification[C]//10th International Conference on Pattern Recognition and Information Processing,Minsk,Belarus,2009:140-145.