Medical volume enhancement using 3-d brushlet transform


Creative Commons License

SELVER M. A., DİCLE O.

JOURNAL OF THE FACULTY OF ENGINEERING AND ARCHITECTURE OF GAZI UNIVERSITY, vol.33, no.4, pp.1215-1230, 2018 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 33 Issue: 4
  • Publication Date: 2018
  • Doi Number: 10.17341/gazimmfd.416421
  • Journal Name: JOURNAL OF THE FACULTY OF ENGINEERING AND ARCHITECTURE OF GAZI UNIVERSITY
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, TR DİZİN (ULAKBİM)
  • Page Numbers: pp.1215-1230
  • Keywords: 3-D medical imaging, transfer function specification, brushlet transform, support vector machines, SEGMENTATION, CLASSIFICATION, REPRESENTATION, STORAGE, IMAGES, TOOL
  • Dokuz Eylül University Affiliated: Yes

Abstract

TF specification controls the visual illustration of medical volumetric data by mapping data values to color and opacity and it is an integrated part of interactive Direct Volume Rendering (DVR). In recent years, the importance of generating multi dimensional domains representing the texture properties has been emphasized in several studies. Accordingly, the superior performance of the brushlet based TF design method and its effective use in 3D visualization is reported in comparison with other statistical or space-frequency based methods (such as wavelet transform, Gabor filter banks, directional filters etc.). This previously developed method uses only radiologist selected Space-Frequency Blocks (SFBs), which are produced by the brushlet transform of 3D medical image series, for reconstruction. The results of its application showed enhanced visualization capabilities especially for the abdominal organs (i.e. liver, kidneys and spleen). In this study, instead of selecting some of the SFBs for reconstruction, a new strategy is proposed to use all SFBs, which are optimally weighted based on the desired 3D image. In accordance with this plan, first s a novel TF specification method, which relies on performing reconstruction with optimal SFB weights, is developed. The optimal SFB weights are calculated through support vector machines in order to minimize the error obtained by the comparison of the weighted SFB reconstruction and the desired 3D visualization. The results obtained by the application of the proposed method to a diverse set of medical image series show improved representation and visualization capabilities compared to SFB selection strategy and manually delineated ground truth.