Open-source algorithm and software for computed tomography-based virtual pancreatoscopy and other applications

Abstract

Pancreatoscopy plays a significant role in the diagnosis and treatment of pancreatic diseases. However, the risk of pancreatoscopy is remarkably greater than that of other endoscopic procedures, such as gastroscopy and bronchoscopy, owing to its severe invasiveness. In comparison, virtual pancreatoscopy (VP) has shown notable advantages. However, because of the low resolution of current computed tomography (CT) technology and the small diameter of the pancreatic duct, VP has limited clinical use. In this study, an optimal path algorithm and super-resolution technique are investigated for the development of an open-source software platform for VP based on 3D Slicer. The proposed segmentation of the pancreatic duct from the abdominal CT images reached an average Dice coefficient of 0.85 with a standard deviation of 0.04. Owing to the excellent segmentation performance, a fly-through visualization of both the inside and outside of the duct was successfully reconstructed, thereby demonstrating the feasibility of VP. In addition, a quantitative analysis of the wall thickness and topology of the duct provides more insight into pancreatic diseases than a fly-through visualization. The entire VP system developed in this study is available at https://github.com/gaoyi/VirtualEndoscopy.git .

Keywords: 3D Slicer; Pancreatic cancer; Pancreatic duct segmentation; Virtual pancreatoscopy.

Fig. 1

Algorithm diagram: a 3D model of the pancreas. Two points (F-1 and F-2) are first set at the b head and c tail of the duct, respectively. d The centerline is then automatically calculated and lumen segmentation is applied

Fig. 2

Software overview. After inputting the 3D CT image and setting the points, the software will output the segmented lumen and generate a VE animation, allowing a statistical analysis to be conducted

Fig. 3

UI of the proposed software

Fig. 4

Use of the Fiducial marker tool to pin-point an area on the slices. From left to right: Slicer modules and the head and tail of the pancreatic duct. The red circle indicates the Fiducial marker tool, and the blue circles indicate the points (point F-1 is at the head of the duct and F-2 is at the tail)

Fig. 5

Results of centerline extraction, in which the first row represents the 3D model and the second row denotes the network topology

Fig. 6

Segmentation results shown in 2D views: a, c, e, g, i, and k are the original images, and b, d, f, h, j and l show the magnified views of the segmentation contour. The results of the proposed method are marked in blue, and those of the manual segmentation are marked in yellow

Fig. 7

Segmentation results shown in Fig. 6 displayed in a 3D view. The Dice coefficients of the first, second, and third row examples are 0.90, 0.87, and 0.84, respectively

Fig. 8

Results of simulated fly-through of the lumen

Fig. 9

Results of curvature of the lumen. a Pancreatic duct of healthy subject; b Pancreatic duct with a simulated cyst (white arrow)