Value of texture analysis based on spectral CT in predicting preoperative lymph node metastasis of gastric cancer

ObjectiveTo study the value of spectral CT texture analysis in predicting preoperative lymph node metastasis in patients with gastric cancer.

MethodsEighty patients(including 57 cases in training group and 23 cases in verification group) with gastric cancer confirmed by surgical resection and pathology were retrospectively analyzed.The special texture analysis software AK was used to segment the lesions and extract the imaging features on the preoperative spectral CT 70 keV venous phase cross-sectional images.The Mann-Whitney U test was used to analyze the features between two groups, and the P < 0.05 features were preserved.The discriminative features were further found by single factor logistic regression analysis.The minimum redundancy maximum correlation method(MRMR) was used to eliminate the 10 features with the highest correlation with the label.Stepwise multiple logistic regression was used to construct the prediction model and final model.The performance of the model was evaluated using ROC analysis.

ResultsIn terms of texture features, the 10 radiology-related features selected by MRMR had better discriminative ability for training group and verification group(AUC>0.64), and the AUC of multivariate logistic regression prediction model was 0.79(0.69-0.89).

ConclusionsThe texture analysis based on spectral CT is expected to be a non-invasive tool for predicting preoperative lymph node metastasis in patients with gastric cancer.