In order to determine ADC, a region of interest (ROI) was established on the ADC map above the required area (prostatic neoplasm, lymph node or normal tissue) with the lowest ADC value identified as the zone with the largest hypo-intensity of MR signal. Taking into consideration that DWIs are morpho-functional images with limited morphological information, in order to improve the accuracy of anatomical comparisons of suspicious areas, we additionally performed mutual overlap of axial T2-WIs and DWIs using RadiAnt DICOM Viewer 2020.2.3 software package, obtaining a color map, where color intensities corresponded to the degrees of diffusion restriction, allowing for accurate spotting of the abnormal lesions detected on DWIs. To obtain the precise ROI position above the area of the lymph node analyzed on ADC maps, we copied the ROI from the respective slice of axial T1-WIs or T2-WIs, which have served as a precise anatomical landmark. In addition to that, to ensure more precise identification of lymph nodes, we proposed a DWIbased method of pelvic lymph node mapping using a maximum intensity projection algorithm, which facilitated spatial identification of lymph nodes and preoperative preparation (Fig. 1). No significant differences were observed when comparing mean sizes of N+ and N– pelvic lymph nodes (p > 0.05). At the same time, when comparing  mean ADC values for N+ and N– pelvic lymph nodes, we did observe a  tatistically significant difference: in metastatic lymph node involvement, this value was 0.74 ± 0.09 × 10-3 mm2/s, while in lymph nodes without  metastatic involvement this value was 1.05 ± 0.23 • 10-3 mm2/s (p < 0.001). Such findings reflect diffusion restriction of hydrogen molecules in N+ lymph nodes due to the increased cellular density in their tissues, which is the case in the development of malignant tumors (Fig. 2). The ROC-analysis using the ADC of DWI MRI for differentiation of N+ and N– pelvic lymph nodes in PCa has shown that in a threshold cut-off value of 0.87 •
10-3 mm2/s, the sensitivity and specificity were 87% and 75%, respectively, with a high accuracy of the method, area under the curve = 0.933; 95% confidence interval (CI) = 0.852–1.0; p < 0.001 (Fig. 3)