Identifying Periampullary Regions in MRI Images Using Deep Learning.

Yong Tang,Weijia Wang,Song Su,Qingxi Guo,Jian Shu,Xinpei Chen,Jiali Wu,Yingjun Zheng

doi:10.3389/fonc.2021.674579

Abstract

BackgroundDevelopment and validation of a deep learning method to automatically segment the peri-ampullary (PA) region in magnetic resonance imaging (MRI) images.MethodsA group of patients with or without periampullary carcinoma (PAC) was included. The PA regions were manually annotated in MRI images by experts. Patients were randomly divided into one training set, one validation set, and one test set. Deep learning methods were developed to automatically segment the PA region in MRI images. The segmentation performance of the methods was compared in the validation set. The model with the highest intersection over union (IoU) was evaluated in the test set.ResultsThe deep learning algorithm achieved optimal accuracies in the segmentation of the PA regions in both T1 and T2 MRI images. The value of the IoU was 0.68, 0.68, and 0.64 for T1, T2, and combination of T1 and T2 images, respectively.ConclusionsDeep learning algorithm is promising with accuracies of concordance with manual human assessment in segmentation of the PA region in MRI images. This automated non-invasive method helps clinicians to identify and locate the PA region using preoperative MRI scanning.

Highlights

The peri-ampulla (PA) region refers to the area within 2cm of the main papilla of the duodenum, including Vater ampulla, lower segment of common bile duct, opening of pancreatic duct, duodenal papilla and duodenal mucosa nearby [1,2,3,4]
The PA region was prone to a series of diseases, including malignant tumors such as periampullary carcinoma (PAC) and benign lesions such as chronic mass pancreatitis, the inflammatory stricture of the lower of common bile duct, or the lower of common bile duct stone etc. [5, 6]
Among all these modern imaging techniques, magnetic resonance imaging (MRI) is a preferable choice to detect the diseases of the PA region for its advantages of excellent softtissue contrast and fewer radiation exposures [5, 7]

Summary

Introduction

The peri-ampulla (PA) region refers to the area within 2cm of the main papilla of the duodenum, including Vater ampulla, lower segment of common bile duct, opening of pancreatic duct, duodenal papilla and duodenal mucosa nearby [1,2,3,4]. This region was deep and narrow in the abdomen and has many adjacent organs and blood vessels, so it is difficult to identify this area using conventional imaging examinations. Development and validation of a deep learning method to automatically segment the peri-ampullary (PA) region in magnetic resonance imaging (MRI) images

Methods

Results

Conclusion