Two abdominal radiologists with 24 and 12 years of experience independently evaluated the DCE-MRI image quality. They reviewed the images using cine mode, with all images from the same section being presented sequentially. Five items (pancreas edge sharpness, motion artifact, streak artifact, noise, and overall image quality) were evaluated with a five-point scale. A score of 1 indicated the most severe image degradation or the worst image quality, and a score of 5 indicated the fewest artifacts/least noise or the best image quality. Pancreas edge sharpness was evaluated on the section that covered the largest area of the pancreas. For the determination of the motion artifact, the presence of multiple lines parallel to the abdominal wall that caused the blurring of the abdominal wall was evaluated. Streak artifact usually appeared as multiple radial lines around the very bright structures or structures outside the field of view. Two radiologists reviewed all of the available images in consensus to determine whether a focal lesion was in the pancreas and made a diagnosis for the focal lesion.
Image processing and analysis for the DCE-MRI was performed using an application for evaluating DCE-MRI (MR Tissue4D) based on commercial software (Syngo.via VB30A, Siemens Healthineers). The perfusion maps were generated using a population-based arterial input function within a sphere-shaped volume of interest containing the entire pancreas and adjacent vessels. Although the software application provided parametric maps based on the Tofts model, we only used measurements from the time–intensity curve. One abdominal radiologist with 12 years of experience in pancreatic MRI performed the image analysis and measured the pancreatic duct diameter in the head, body, and tail of the pancreas. The radiologist also drew six regions of interest (ROIs) in three areas of the vessels (the descending aorta at the left crus level, celiac axis, and superior mesenteric artery (SMA)) and three areas of the pancreas (head, body, and tail). The demarcation of the head, body, and tail of the pancreas was based on the 8th edition of the American Joint Committee on Cancer (AJCC) system [16 ], as follows: the head is to the right of the superior mesenteric–portal vein confluence, the body is between the left border of the superior mesenteric vein and the left border of the aorta, and the tail is between the left border of the aorta and the hilum of the spleen. ROIs in the vessels were free-hand drawn as large as possible while avoiding the vessel wall, and those in the pancreas had sizes larger than 50 mm2. From each ROI, we measured the peak-enhancement time, which is the time between the start of image acquisition and the highest signal intensity. The delay time was based on the time elapsed between the peak-enhancement time in the aorta and the pancreas. The peak concentration in the ROI was recorded from the time–intensity curve. Figure 1 depicts the concepts of the investigated parameters (Figure 1).
Free full text: Click here