Prediction of different stages of rectal cancer: Texture analysis based on diffusion-weighted images and apparent diffusion coefficient maps

Figure 1 Flow chart adopted in this study. MRI: Magnetic resonance imaging; DWI: Diffusion-weighted imaging; NAT: Neoadjuvant chemoradiotherapy; ADC: Apparent diffusion coefficient; ROI: Region of interest.

Figure 2 The region of interest segmentation results for a randomly selected case. A-C: Represent diffusion-weighted imaging (DWI)_b=0, DWI_b=1000 and the apparent diffusion coefficient image on the same slice, respectively; D-F: With regard to T2WI and DWI, the lesion region of interest was drawn on the apparent diffusion coefficient map (F) and copied onto the DWI_b=0 (D) and DWI_b=1000 (E) images.

Figure 3 Receiver operating characteristic curves obtained with different discriminatory models for predicting T1-2 and T3-4 stage tumors. AUC: Area under the receiver operating characteristic curve; ADC: Apparent diffusion coefficient.

Figure 4 Receiver operating characteristic curves obtained with different discriminatory models for predicting N0 and N1-2 stage tumors. AUC: Area under the receiver operating characteristic curve; ADC: Apparent diffusion coefficient.