Download PDFOpen PDF in browser

Deep Learning based Segmentation of Lumbar Vertebrae from CT Images

4 pagesPublished: July 12, 2018

Abstract

We present a method to address the challenging problem of automatic segmentation of lumbar vertebrae from CT images acquired with varying fields of view. Our method is based on cascaded 3D Fully Convolutional Networks (FCNs) consisting of a localization FCN and a segmentation FCN. More specifically, in the first step we train a regression 3D FCN (we call it “LocalizationNet”) to find the bounding box of the lumbar region. After that, a 3D U-net like FCN (we call it “SegmentationNet”) is then developed, which after training, can perform a pixel-wise multi-class segmentation to map a cropped lumber region volumetric data to its volume-wise labels. Evaluated on publicly available datasets, our method achieved an average Dice coefficient of 95.77 ± 0.81% and an average symmetric surface distance of 0.37 ± 0.06 mm.

Keyphrases: ct, deep learning, segmentation, spine

In: Wei Tian and Ferdinando Rodriguez Y Baena (editors). CAOS 2018. The 18th Annual Meeting of the International Society for Computer Assisted Orthopaedic Surgery, vol 2, pages 94-97.

BibTeX entry
@inproceedings{CAOS2018:Deep_Learning_based_Segmentation,
  author    = {Rens Janssens and Guoyan Zheng},
  title     = {Deep Learning based Segmentation of Lumbar Vertebrae from CT Images},
  booktitle = {CAOS 2018. The 18th Annual Meeting of the International Society for Computer Assisted Orthopaedic Surgery},
  editor    = {Wei Tian and Ferdinando Rodriguez Y Baena},
  series    = {EPiC Series in Health Sciences},
  volume    = {2},
  publisher = {EasyChair},
  bibsource = {EasyChair, https://easychair.org},
  issn      = {2398-5305},
  url       = {/publications/paper/8cHM},
  doi       = {10.29007/vt7v},
  pages     = {94-97},
  year      = {2018}}
Download PDFOpen PDF in browser