Abstract:Purpose: MRI guided radiotherapy is particularly attractive for abdominal targets with low CT contrast. To fully utilize this modality for pancreas tracking, automated segmentation tools are needed. A hybrid gradient, region growth and shape constraint (hGReS) method to segment 2D upper abdominal dynamic MRI is developed for this purpose. Methods: 2D coronal dynamic MR images of 2 healthy volunteers were acquired with a frame rate of 5 f/second. The regions of interest (ROIs) included the liver, pancreas and stomach. The first frame was used as the source where the centers of the ROIs were annotated. These center locations were propagated to the next dynamic MRI frame. 4-neighborhood region transfer growth was performed from these initial seeds for rough segmentation. To improve the results, gradient, edge and shape constraints were applied to the ROIs before final refinement using morphological operations. Results from hGReS and 3 other automated segmentation methods using edge detection, region growth and level set were compared to manual contouring. Results: For the first patient, hGReS resulted in the organ segmentation accuracy as measure by the Dices index (0.77) for the pancreas. The accuracy was slightly superior to the level set method (0.72), and both are significantly more accuratemore » than the edge detection (0.53) and region growth methods (0.42). For the second healthy volunteer, hGReS reliably segmented the pancreatic region, achieving a Dices index of 0.82, 0.92 and 0.93 for the pancreas, stomach and liver, respectively, comparing to manual segmentation. Motion trajectories derived from the hGReS, level set and manual segmentation methods showed high correlation to respiratory motion calculated using a lung blood vessel as the reference while the other two methods showed substantial motion tracking errors. hGReS was 10 times faster than level set. Conclusion: We have shown the feasibility of automated segmentation of the pancreas anatomy based on dynamic MRI.« less

Automated pancreas segmentation from three-dimensional contrast-enhanced computed tomography

Harnessing 2D Networks and 3D Features for Automated Pancreas Segmentation from Volumetric CT Images

Automated Pancreas Segmentation Using Recurrent Adversarial Learning

Fully Automated Pancreas Segmentation with Two-stage 3D Convolutional Neural Networks

A Bottom-Up Approach for Automatic Pancreas Segmentation in Abdominal CT Scans

Automatic Segmentation Algorithm of Pancreatic CT Images Based on 2-3D Mixed Convolutional Neural Network

Regression Forest-Based Atlas Localization and Direction Specific Atlas Generation for Pancreas Segmentation

3D FCN Feature Driven Regression Forest-Based Pancreas Localization and Segmentation

EP-1887: Automated 3D MRI Pancreas Segmentation

Multi-Contrast Computed Tomography Atlas of Healthy Pancreas

DRAN: Deep Recurrent Adversarial Network for Automated Pancreas Segmentation

Multi-atlas Pancreas Segmentation: Atlas Selection Based on Vessel Structure

Towards dense volumetric pancreas segmentation in CT using 3D fully convolutional networks

Automated CT pancreas segmentation for acute pancreatitis patients by combining a novel object detection approach and U-Net

Feasibility of automated pancreas segmentation based on dynamic MRI.

nnU-Net-Based Pancreas Segmentation and Volume Measurement on CT Imaging in Patients with Pancreatic Cancer

SU-E-J-168: Automated Pancreas Segmentation Based On Dynamic MRI

Iterative 3D Feature Enhancement Network for Pancreas Segmentation from CT Images

Feasibility of automated 3-dimensional magnetic resonance imaging pancreas segmentation.

MCMC Guided CNN Training and Segmentation for Pancreas Extraction

SU‐E‐J‐168: Automated Pancreas Segmentation Based on Dynamic MRI