Source-linked AI summary

GBM Volumetry using the 3D Slicer Medical Image Computing Platform

Jan Egger, Tina Kapur, Andriy Fedorov, Steve Pieper, James V. Miller, Harini Veeraraghavan, Bernd Freisleben, Alexandra Golby, Christopher Nimsky, Ron Kikinis

arXiv:1303.0964v1cs.CV

TL;DR

Accurate, repeatable tumor-volume measurement matters for clinical care, while slice-by-slice GBM segmentation is time-consuming. This study evaluates 3D Slicer’s GrowCut segmentation against manual contouring, finding substantially shorter segmentation times with comparable segmentation agreement.

  • Problem

    Accurate and repeatable tumor-volume calculation is important for clinical care, whereas geometric approximations may provide only rough tumor-volume estimates.

  • Method

    The study compares GBM segmentations by four physicians using 3D Slicer’s interactive GrowCut algorithm and manual slice-by-slice contouring across 10 patients.

  • Results

    GrowCut segmentation took about 60% of manual segmentation time, including editing, while the evaluation compared segmentation agreement using Dice Similarity Coefficient and Hausdorff Distance.

  • Takeaways & Limitations

    3D Slicer with GrowCut is presented as a viable alternative for monitoring GBM patients, with computed tumor volumes available for comparison with follow-up scans.

  • Takeaways & Limitations

    Future work is needed to evaluate the method on lower-grade gliomas, whose MRI outlines may rely on surrounding edema rather than contrast-enhancing T1-weighted images.

Abstract

from arXiv · show

Volumetric change in glioblastoma multiforme (GBM) over time is a critical factor in treatment decisions. Typically, the tumor volume is computed on a slice-by-slice basis using MRI scans obtained at regular intervals. (3D)Slicer - a free platform for biomedical research - provides an alternative to this manual slice-by-slice segmentation process, which is significantly faster and requires less user interaction. In this study, 4 physicians segmented GBMs in 10 patients, once using the competitive region-growing based GrowCut segmentation module of Slicer, and once purely by drawing boundaries completely manually on a slice-by-slice basis. Furthermore, we provide a variability analysis for three physicians for 12 GBMs. The time required for GrowCut segmentation was on an average 61% of the time required for a pure manual segmentation. A comparison of Slicer-based segmentation with manual slice-by-slice segmentation resulted in a Dice Similarity Coefficient of 88.43 +/- 5.23% and a Hausdorff Distance of 2.32 +/- 5.23 mm.

Loading 1303.0964v1…