Tech Valley News

Kitware Receives $600K NIH Grant

Kitware, a company that builds open source platforms and develops advanced research solutions, has received a two-year grant from the National Institutes of Health (NIH) totaling more than $600,000 to focus research efforts on developing approach-specific, multi-GPU, multi-tool, high-realism neurosurgery simulation. The goal of the research is to work towards an interactive simulator that replicates future neurosurgery cases of young surgeons, enabling hospitals faced with compressed intern schedules to accelerate training and improve skills.

“Simulation has made limited inroads in neurosurgery despite its proven advantages over traditional surgical training,” said Dr. Michel Audette, R&D Engineer at Kitware. “Our objective is to empower the neurosurgical field with simulation training methods that directly correlate to better, more accurate patient care.”

Kitware has developed a variety of applications and toolkits to support the analysis and visualization of complex biological systems. As leaders in the field of image analysis, registration and segmentation Kitware offers GPU-based methods for data display, volume rendering, image processing and advanced interactive. As a participant in large-scale research initiatives such as the NIH US National Center for Biomedical Computing, Kitware is part of the nation’s core effort to build a computational infrastructure for biomedical computing.

Historically, simulation has over-emphasized needle insertion and failed to address requirements of surgeons-in-training on potential caseloads. Existing simulators do not ably reconcile the conflicting clinical requirements of neurosurgery; their underlying models are insufficiently sparse to resolve large tissue displacements and they are insufficiently descriptive of critical tissues at risk in each intervention.

Kitware will set out to prove that for interactive neurosurgery simulation to be relevant and technically feasible, the following has to occur:
  • Anatomical modeling must be sufficiently descriptive in intra-surgical motion and tissue morphology;
  • Biomechanics must be faithful to tissue response;
  • Haptics must afford multi-tool interaction;
  • And the system must meet clinical requirements reflecting the specific surgical approach and pathology.
Kitware’s Insight Segmentation and Registration Toolkit (ITK) will be used to produce a descriptive and patient-specific application modeling brain anatomy, requiring innovations in segmentation, registration and meshing; VTK for interactive visualization; and expand Kitware’s new Vessel Toolkit. This project will also build on Kitware’s expertise in GPU programming and haptics. Through this proposed work, Kitware will expand its leadership in computer vision towards advancements for photorealistic rendering in surgery simulation, as well as pursue research in medical ontologies.