Analysis of dose heterogeneity using a subvolume-DVH

Research output: Contribution to journalArticle

Abstract

The dose-volume histogram (DVH) is universally used in radiation therapy for its highly efficient way of summarizing three-dimensional dose distributions. An apparent limitation that is inherent to standard histograms is the loss of spatial information, e.g. it is no longer possible to tell where low- and high-dose regions are, and whether they are connected or disjoint. Two methods for overcoming the spatial fragmentation of low- and high-dose regions are presented, both based on the gray-level size zone matrix, which is a two-dimensional histogram describing the frequencies of connected regions of similar intensities. The first approach is a quantitative metric which can be likened to a homogeneity index. The large cold spot metric (LCS) is here defined to emphasize large contiguous regions receiving too low a dose; emphasis is put on both size, and deviation from the prescribed dose. In contrast, the subvolume-DVH (sDVH) is an extension to the standard DVH and allows for a qualitative evaluation of the degree of dose heterogeneity. The information retained from the two-dimensional histogram is overlaid on top of the DVH and the two are presented simultaneously. Both methods gauge the underlying heterogeneity in ways that the DVH alone cannot, and both have their own merits - the sDVH being more intuitive and the LCS being quantitative.

Details

Authors
Organisations
External organisations
  • Skåne University Hospital
  • Lund University
Research areas and keywords

Subject classification (UKÄ) – MANDATORY

  • Radiology, Nuclear Medicine and Medical Imaging

Keywords

  • dose inhomogeneity, DVH, GLSZM, texture analysis, treatment plan comparison
Original languageEnglish
Pages (from-to)N517-N524
JournalPhysics in Medicine and Biology
Volume62
Issue number21
Publication statusPublished - 2017 Oct 19
Publication categoryResearch
Peer-reviewedYes