This study was performed to assess whether iterative reconstruction can reduce radiation dose while maintaining acceptable image quality, and to investigate whether perfusion parameters vary from conventional filtered back projection (FBP) at the low-tube-voltage (80-kVp) during whole-pancreas perfusion examination using a 256-slice CT. 76 patients with known or suspected pancreatic mass underwent whole-pancreas perfusion by a 256-slice CT. High- and low-tube-voltage CT images were acquired. 120-kVp image data (protocol A) and 80-kVp image data (protocol B) were reconstructed with conventional FBP, and 80-kVp image data were reconstructed with iDose⁴ (protocol C) iterative reconstruction. The image noise; contrast-to-noise ratio (CNR) relative to muscle for the pancreas, liver, and aorta; and radiation dose of each protocol were assessed quantitatively. Overall image quality was assessed qualitatively. Among 76 patients, 23 were eventually proven to have a normal pancreas. Perfusion parameters of normal pancreas in each protocol including blood volume, blood flow, and permeability-surface area product were measured. In the quantitative study, protocol C reduced image noise by 36.8% compared to protocol B (P<0.001). Protocol C yielded significantly higher CNR relative to muscle for the aorta, pancreas and liver compared to protocol B (P<0.001), and offered no significant difference compared to protocol A. In the qualitative study, protocols C and A gained similar scores and protocol B gained the lowest score for overall image quality (P<0.001). Mean effective doses were 23.37 mSv for protocol A and 10.81 mSv for protocols B and C. There were no significant differences in the normal pancreas perfusion values among three different protocols. Low-tube-voltage and iDose⁴ iterative reconstruction can dramatically decrease the radiation dose with acceptable image quality during whole-pancreas CT perfusion and have no significant impact on the perfusion parameters of normal pancreas compared to the conventional FBP reconstruction using a 256-slice CT scanner.

Anastomosing haemangioma (AH) is a recently described, unusual variant of capillary hemangioma that appears to be unique to the genitourinary system, with a particular proclivity for the kidney. AH is a subtype of capillary haemangioma, which is rarely encountered in clinical practice, particularly in the liver. We herein present the case of a 57‑year‑old woman with an incidental finding on magnetic resonance imaging of a local lesion in the liver, sized 3.3x3.0 cm. The patient underwent hepatectomy with a good postoperative recovery. The histopathological diagnosis was AH of the liver. To the best of our knowledge, this is the first case report of hepatic AH.

Retrospective assessment was performed of the minimum apparent diffusion coefficient (minADC) together with quantitative assessment of the histopathological features of resected tumour tissue and the Ki-67 labelling index (LI) for predicting pineal parenchymal tumour (PPT) grade. Magnetic resonance images of tumours from 26 male and female patients (mean age 27.7 years) with pathologically confirmed PPTs, comprising 10 pineocytomas (PCs), seven pineal parenchymal tumours of intermediate differentiation (PPTIDs) and nine pineoblastomas (PBs), were reviewed retrospectively. Using ADC maps derived from isotropic diffusion-weighted images, the minADC value of each tumour was determined postoperatively from several regions of interest defined in the tumour, excluding cystic, necrotic, haemorrhagic or calcified components. Surgical intervention was undertaken in all cases. The Ki-67 LI was measured in surgical specimens. Mean minADC and Ki-67 LI showed significant differences among PCs, PPTIDs and PBs, and minADC was negatively correlated with Ki-67 LI. It is concluded that the minADC of PPTs provide useful additional information when predicting tumour grading.