Information

Abstract

Neurosurgeons make decisions based on expert knowledge that takes factors such as safety margins, the avoidance of risk structures, trajectory length and trajectory angle into consideration. While some of those factors are mandatory, others can be optimized in order to obtain the best possible trajectory under the given circumstances. Through comparison with the actually chosen trajectories from real biopsies and qualitative interviews with domain experts, we identified important rules for trajectory planning. In this thesis, we present BrainXplore, an interactive visual analysis tool for aiding neurosurgeons in planning brain biopsies. BrainXplore is an extendable Biopsy Planning framework that incorporates those rules while at the same time leaving full flexibility for their customization and adding of new structures at risk. Automatically computed candidate trajectories can be incrementally refined in an interactive manner until an optimal trajectory is found. We employ a spatial index server as part of our system that allows us to access distance information on an unlimited number of risk structures at arbitrary resolution. Furthermore, we implemented InfoVis techniques such as Parallel Coordinates and risk signature charts to drive the decision process. As a case study, BrainXPlore offers a variety of information visualization modalities to present multivariate data in different ways. We evaluated BrainXPlore on a real dataset and accomplished acceptable results. The participating neurosurgeon gave us the feedback that BrainXPlore can decrease the time needed for biopsy planning and aid novice users in their decision making process.

Additional Files and Images

Additional images and videos

Additional files

Weblinks

No further information available.

BibTeX

@mastersthesis{Pezenka-2016-MT,
  title =      "BrainXPlore - Decision finding in Brain Biopsy Planning",
  author =     "Lukas Pezenka",
  year =       "2017",
  abstract =   "Neurosurgeons make decisions based on expert knowledge that
               takes factors such as safety margins, the avoidance of risk
               structures, trajectory length and trajectory angle into
               consideration. While some of those factors are mandatory,
               others can be optimized in order to obtain the best possible
               trajectory under the given circumstances. Through comparison
               with the actually chosen trajectories from real biopsies and
               qualitative interviews with domain experts, we identified
               important rules for trajectory planning. In this thesis, we
               present BrainXplore, an interactive visual analysis tool for
               aiding neurosurgeons in planning brain biopsies. BrainXplore
               is an extendable Biopsy Planning framework that incorporates
               those rules while at the same time leaving full flexibility
               for their customization and adding of new structures at
               risk. Automatically computed candidate trajectories can be
               incrementally refined in an interactive manner until an
               optimal trajectory is found. We employ a spatial index
               server as part of our system that allows us to access
               distance information on an unlimited number of risk
               structures at arbitrary resolution. Furthermore, we
               implemented InfoVis techniques such as Parallel Coordinates
               and risk signature charts to drive the decision process. As
               a case study, BrainXPlore offers a variety of information
               visualization modalities to present multivariate data in
               different ways. We evaluated BrainXPlore on a real dataset
               and accomplished acceptable results. The participating
               neurosurgeon gave us the feedback that BrainXPlore can
               decrease the time needed for biopsy planning and aid novice
               users in their decision making process.",
  month =      dec,
  address =    "Favoritenstrasse 9-11/E193-02, A-1040 Vienna, Austria",
  school =     "Institute of Computer Graphics and Algorithms, Vienna
               University of Technology ",
  URL =        "https://www.cg.tuwien.ac.at/research/publications/2017/Pezenka-2016-MT/",
}