Format

Send to

Choose Destination
Cancer Imaging. 2018 Oct 17;18(1):37. doi: 10.1186/s40644-018-0170-8.

The differentiation of pancreatic neuroendocrine carcinoma from pancreatic ductal adenocarcinoma: the values of CT imaging features and texture analysis.

Author information

1
Department of Radiology, the First Affiliated Hospital, College of Medicine Zhejiang University, 79 Qingchun road, Hangzhou, 310003, China.
2
Department of Laboratory Medicine, the First Affiliated Hospital, College of Medicine Zhejiang University, 79 Qingchun road, Hangzhou, 310003, China.
3
Department of Radiology, the Affiliated Hospital of Nanjing University of Chinese Medicine, 155 Hanzhong road, Nanjing, 210029, China.
4
Department of Radiology, the First Affiliated Hospital, College of Medicine Zhejiang University, 79 Qingchun road, Hangzhou, 310003, China. gerxyuan@zju.edu.cn.
5
Department of Radiology, the Affiliated Hospital of Nanjing University of Chinese Medicine, 155 Hanzhong road, Nanjing, 210029, China. chxwin@163.com.

Abstract

BACKGROUND:

Imaging findings for pancreatic neuroendocrine carcinoma (PNEC) and pancreatic ductal adenocarcinoma (PDAC) often overlap. The aim of this study was to demonstrate the value of computed tomography (CT) imaging features and texture analysis to differentiate PNEC from PDAC.

METHODS:

Twenty-eight patients with pathologically-proved PDAC and 14 patients with PNEC were included in this study. CT imaging findings, including tumor boundary, size, enhancement degree, duct dilatation and parenchymal atrophy were used to compare PDAC and PNEC. CT texture features were extracted from CT images at the arterial and portal phases.

RESULTS:

More PNEC than PDAC had well-defined margins (57.1% vs 25.0%, p = 0.04). Parenchymal atrophy was more common in PDAC than in PNEC (67.9% vs 28.1%, p = 0.02). CT attenuation values (HU) and contrast ratios of PNEC inthe arterial and portal phases were higher than those of PDAC (p < 0.05 or 0.01). Entropy was lower and uniformity was higher in PNEC compare to PDAC at the arterial phase (p < 0.05). Contrast ratio showed the highest area under curve (AUC) for differentiating PNEC from PDAC (AUC = 0.98-0.99). Entropy and uniformity also showed an acceptable AUC (0.71-0.72).

CONCLUSIONS:

Our data indicate that CT imaging features, including tumor margin, enhanced degree and parenchymal atrophy, as well as texture parameters can aid in the differentiation of PNEC from PDAC.

KEYWORDS:

Computed tomography.; Pancreatic ductal adenocarcinoma.; Pancreatic neuroendocrine carcinoma; Texture analysis.

PMID:
30333055
PMCID:
PMC6192319
DOI:
10.1186/s40644-018-0170-8
[Indexed for MEDLINE]
Free PMC Article

Supplemental Content

Full text links

Icon for BioMed Central Icon for PubMed Central
Loading ...
Support Center