Resectable pancreatic ductal adenocarcinoma: association between preoperative CT texture features and metastatic nodal involvement

Abstract

Background: To explore the relationship between the lymph node status and preoperative computed tomography images texture features in pancreatic cancer.

Methods: A total of 155 operable pancreatic cancer patients (104 men, 51 women; mean age 63.8 ± 9.6 years), who had undergone contrast-enhanced computed tomography in the arterial and portal venous phases, were enrolled in this retrospective study. There were 73 patients with lymph node metastases and 82 patients without nodal involvement. Four different data sets, with thin (1.25 mm) and thick (5 mm) slices (at arterial phase and portal venous phase) were analysed. Texture analysis was performed by using MaZda software. A combination of feature selection algorithms was used to determine 30 texture features with the optimal discriminative performance for differentiation between lymph node positive and negative groups. The prediction performance of the selected feature was evaluated by receiver operating characteristic (ROC) curve analysis.

Results: There were 10 texture features with significant differences between two groups and significance in ROC analysis were identified. They were WavEnLH_s-2(wavelet energy with rows and columns are filtered with low pass and high pass frequency bands with scale factors 2) from wavelet-based features, 135dr_LngREmph (long run emphasis in 135 direction) and 135dr_Fraction (fraction of image in runs in 135 direction) from run length matrix-based features, and seven variables of sum average from coocurrence matrix-based features (SumAverg). The ideal cutoff value for predicting lymph node metastases was 270 for WavEnLH_s-2 (positive likelihood ratio 2.08). In addition, 135dr_LngREmph and 135dr_Fraction were correlated with the ratio of metastatic to examined lymph nodes.

Conclusions: Preoperative computed tomography high order texture features provide a useful imaging signature for the prediction of nodal involvement in pancreatic cancer.

Keywords: Computed tomography; Computer-assisted image processing; Metastases; Pancreatic ductal adenocarcinoma; Texture analysis.

Fig. 1

Illustration of flow diagram of included patients

Fig. 2

Illustration of drawing the region of interest (ROI) by MaZda software 77 years old male, pancreatic ductal adenocarcinoma (PDAC) in the pancreatic tail (green ROI), without lymph node (LN) metastases, axial virtual monochromatic (VMC) CT images at 65 k electron voltage (keV) on late arterial phase with 5 mm slice thickness (a, b). 53 years old male, PDAC in the pancreatic head (red ROI), with 1/25 metastatic LN, axial VMC CT images at 60 keV on portal venous phase with 1.25 mm slice thickness (c, d)

Fig. 3

Receiver operating characteristic (ROC) curve of wavelet transform based feature WavEnLH_s-2 (area under ROC curve = 0.630, 95% confidence interval 0.549 to 0.706) from the virtual monochromatic CT image at 60 k electron voltage (keV) on late arterial phase with 1.25 mm slice thickness for differentiating lymph node metastases positive and negative groups of pancreatic ductal adenocarcinoma (155 cases)

Fig. 4

Correlations of ratio of metastatic to examined lymph nodes (LNR) with run-length matrix based features 135dr_LngREmph (a) and 135dr_Fraction (b) from virtual monochromatic CT images at 65 k electron voltage (keV) on late arterial phase (AP) with 5 mm slice thickness in the pancreatic ductal adenocarcinoma patients (155 cases), analyzed using the non-parametric Spearman’s correlation coefficients method