Predictive Value of CT Texture Parameters for Postoperative Recurrence in Patients with Colorectal Cancer
Objective To investigate the value of CT texture parameters in predicting postoperative re-currence of colorectal cancer(CRC)patients.Methods The data of 70 CRC patients who received lapa-roscopic total mesenterectomy in our hospital from October 2021 to December 2022 and underwent en-hanced CT examination before operation were retrospectively analyzed.The enhanced CT images(arterial and venous phases)were analyzed for texture analysis,and the relevant texture parameters were extracted.According to the 1-year follow-up results,patients in the two groups were divided into recurrence group and non-recurrence group.The texture parameters of the two groups were compared.The risk factors con-cerned with postoperative recurrence of CRC patients were analyzed.The receiver operating characteristic(ROC)curve was drawn,and the area under the curve(AUC)was used to evaluate the predictive value of CT texture parameters on postoperative recurrence.Results Among the 70 patients,16 cases(22.86％)recurred one year after operation,and 54 cases(77.14％)had no recurrence.The arterial phase entropy,ve-nous phase inhomogeneity and venous phase entropy in the recurrent group were higher than those in the non-recurrent group(P＜0.05).Multivariate analysis showed that arterial phase entropy(OR=5.109,95％CI 1.354-8.864),venous phase inhomogeneity(OR=4.943,95％CI 1.235-8.661),venous phase en-tropy(OR=5.171,95％CI 1.554～8.987)were risk factors for postoperative recurrence of CRC patients(P＜0.05).ROC curve analysis showed that arterial phase entropy(AUC=0.768),venous phase inho-mogeneity(AUC=0.702)and venous phase entropy(AUC=0.873)all had certain accuracy in predicting postoperative recurrence of CRC patients,and the combination of the three had higher predictive efficacy.Sensitivity,specificity and AUC were 80.25％,98.20％and 0.921,respectively.Conclusion The combi-nation of arterial phase entropy,venous phase inhomogeneity and entropy of CT texture parameters has certain predictive value for postoperative recurrence of CRC patients.
NETL
NSTL
万方数据
