Visual analytics for the exploration of multiparametric cancer imaging

Renata Raidou, Marta Paes Moreira, Wouter van Elmpt, Marcel Breeuwer, Anna Vilanova
Visual analytics for the exploration of multiparametric cancer imaging
In Visual Analytics Science and Technology (VAST), 2014 IEEE Conference on Visualization, (), 2014.

Information

Abstract

Tumor tissue characterization can play an important role in thediagnosis and design of effective treatment strategies. In orderto gather and combine the necessary tissue information, multi-modal imaging is used to derive a number of parameters indica-tive of tissue properties. The exploration and analysis of relation-ships between parameters and, especially, of differences among dis-tinct intra-tumor regions is particularly interesting for clinical re-searchers to individualize tumor treatment. However, due to highdata dimensionality and complexity, the current clinical workflowis time demanding and does not provide the necessary intra-tumorinsight. We implemented a new application for the exploration ofthe relationships between parameters and heterogeneity within tu-mors. In our approach, we employ a well-known dimensionalityreduction technique [5] to map the high-dimensional space of tis-sue properties into a 2D information space that can be interactivelyexplored with integrated information visualization techniques. Weconducted several usage scenarios with real-patient data, of whichwe present a case of advanced cervical cancer. First indicationsshow that our application introduces new features and functionali-ties that are not available within the current clinical approach.

Additional Files and Images

Weblinks

No further information available.

BibTeX

@article{raidou_vis14,
  title =      "Visual analytics for the exploration of multiparametric
               cancer imaging",
  author =     "Renata Raidou and Marta Paes Moreira and Wouter van Elmpt
               and Marcel Breeuwer and Anna Vilanova",
  year =       "2014",
  abstract =   "Tumor  tissue  characterization  can  play  an  important 
               role  in  thediagnosis  and  design  of  effective 
               treatment  strategies.    In  orderto  gather  and  combine 
               the  necessary  tissue  information,  multi-modal  imaging 
               is  used  to  derive  a  number  of  parameters  indica-tive
               of tissue properties.  The exploration and analysis of
               relation-ships between parameters and, especially, of
               differences among dis-tinct intra-tumor regions is
               particularly interesting for clinical re-searchers to
               individualize tumor treatment.  However, due to highdata
               dimensionality and complexity, the current clinical
               workflowis time demanding and does not provide the necessary
               intra-tumorinsight.  We implemented a new application for
               the exploration ofthe relationships between parameters and
               heterogeneity within tu-mors.   In our approach,  we employ
               a well-known dimensionalityreduction technique [5] to map
               the high-dimensional space of tis-sue properties into a 2D
               information space that can be interactivelyexplored with
               integrated information visualization techniques. 
               Weconducted several usage scenarios with real-patient data,
               of whichwe  present  a  case  of  advanced  cervical 
               cancer.   First  indicationsshow that our application
               introduces new features and functionali-ties that are not
               available within the current clinical approach.",
  journal =    "In Visual Analytics Science and Technology (VAST), 2014 IEEE
               Conference on Visualization",
  pages =      "--",
  URL =        "https://www.cg.tuwien.ac.at/research/publications/2014/raidou_vis14/",
}