Patient-specific semi-supervised learning for postoperative brain tumor segmentation

Med Image Comput Comput Assist Interv. 2014;17(Pt 1):714-21. doi: 10.1007/978-3-319-10404-1_89.

Abstract

In contrast to preoperative brain tumor segmentation, the problem of postoperative brain tumor segmentation has been rarely approached so far. We present a fully-automatic segmentation method using multimodal magnetic resonance image data and patient-specific semi-supervised learning. The idea behind our semi-supervised approach is to effectively fuse information from both pre- and postoperative image data of the same patient to improve segmentation of the postoperative image. We pose image segmentation as a classification problem and solve it by adopting a semi-supervised decision forest. The method is evaluated on a cohort of 10 high-grade glioma patients, with segmentation performance and computation time comparable or superior to a state-of-the-art brain tumor segmentation method. Moreover, our results confirm that the inclusion of preoperative MR images lead to a better performance regarding postoperative brain tumor segmentation.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Algorithms
  • Brain Neoplasms / complications
  • Brain Neoplasms / pathology*
  • Brain Neoplasms / surgery*
  • Diagnosis, Differential
  • Glioma / complications
  • Glioma / pathology*
  • Glioma / surgery*
  • Humans
  • Image Enhancement / methods
  • Image Interpretation, Computer-Assisted / methods*
  • Neurosurgical Procedures / adverse effects
  • Neurosurgical Procedures / methods
  • Pattern Recognition, Automated / methods*
  • Postoperative Care / methods
  • Postoperative Hemorrhage / etiology
  • Postoperative Hemorrhage / pathology*
  • Reproducibility of Results
  • Sensitivity and Specificity
  • Treatment Outcome