Segmentation of pelvic vessels in pediatric MRI using a patch-based deep learning approach

Abstract :

In this paper, we propose a patch-based deep learning ap- proach to segment pelvic vessels in 3D MRI images of pediatric patients. For a given T2 weighted MRI volume, a set of 2D axial patches are extracted using a limited number of user-selected landmarks. In order to take into account the volumetric information, successive 2D axial patches are combined together, producing a set of pseudo RGB color images. These RGB images are then used as input for a convolutional neural network (CNN), pre-trained on the ImageNet dataset, which re- sults into both segmentation and vessel labeling as veins or arteries. The proposed method is evaluated on 35 MRI volumes of pediatric patients, obtaining an average segmentation accuracy in terms of Average Sym- metric Surface Distance of ASSD = 0.89 ± 0.07 mm and Dice Index of DC = 0.79 ± 0.02.

Complete list of metadatas

https://hal.telecom-paristech.fr/hal-02287946
Contributor : Telecomparis Hal <>
Submitted on : Friday, September 13, 2019 - 5:31:25 PM
Last modification on : Thursday, October 17, 2019 - 12:37:00 PM

Identifiers

  • HAL Id : hal-02287946, version 1

Citation

Alessio Virzi, Pietro Gori, Cécile Muller, Eva Mille, Quoc Peyrot, et al.. Segmentation of pelvic vessels in pediatric MRI using a patch-based deep learning approach. PIPPI MICCAI Workshop, 2018, Granada, Spain. pp.97-106. ⟨hal-02287946⟩

Share

Metrics

Record views

13