JMIR Publications

ABSTRACT

Background:

Pancreatic cystic lesions (PCL) are a frequent and underreported incidental finding on computed tomography (CT) scans and can evolve to pancreatic cancer, the most lethal one with less than five months of life expectancy.

Objective:

The aim of this study is to develop and validate an artificial deep neural network (Attention-Gate U-Net; AGNet) for automated detection of pancreatic cystic lesions. This kind of technology can help radiologists to cope with an increasing demand of cross-sectional imaging tests and increase the number of PCLs incidentally detected, therefore increasing the early detection of pancreatic cancer.

Methods:

We adapted and evaluated an algorithm based on an Attention-Gate U-Net architecture for automated detection of PCL on CTs. A total of 335 abdominal CTs with pancreatic cystic lesions and control cases were manually segmented in 3D by two radiologists with over 10 years of experience in consensus with a board-certified radiologist specialized in abdominal radiology. This information was used to train a neural network for segmentation followed by a post-processing pipeline that filtered the results of the network and applied some physical constraints such as the expected position of the pancreas to minimize the number of false positives.

Results:

From 335 studies included in the study, 297 had a pancreatic cystic lesion, including serous cystadenoma (SCA), intraductal pseudopapillary mucinous neoplasia (IPMN), mucinous cystic neoplasm (MCN) and pseudocysts (PCYST). The Shannon Index of the chosen dataset was 0.991 with an evenness of 0.902. The mean sensitivity obtained in the detection of these lesions was 93.1% and the specificity 81.8±0.1%.

Conclusions:

This study shows a good performance of an automated artificial deep neural network in the detection of PCL on both non-contrast and contrast-enhanced abdominal CT scans.