Mixed Reality Image Guidance for Cardiac Interventional Surgery

Principal Investigator: 

Bobak MosadeghAssociate Professor of Biomedical Engineering in Radiology

Background & Unmet Need

  • Minimally invasive, image-guided cardiac interventions are increasingly available as substitutes for more invasive surgical approaches
  • With these minimally invasive procedures, tools for visualization are needed to improve guidance and lower learning curves
  • Current visualization techniques like fluoroscopy are limited to 2D projections, or are unable to give real-time feedback like pre-operative CT/MRI
  • Advanced fusion imaging visualizations still don’t provide quantitative tracking of the catheters in 3D space, and so cannot be used to guide catheter depth or orientation
  • Unmet Need: Real-time, visual guidance systems for cardiac procedures wherein the catheter can be tracked in 3D space using a single fluoroscopic view

Technology Overview

  • The Technology: A novel, mixed reality guidance system which combines holographic representations of the heart and tracking of catheter position in real time
  • A 3D, holographic representation of the heart is generated using preoperative cardiac CT images
  • The catheter is tracked via intra-operative fluoroscopy, and machine learning is used to locate the depth of catheter in 3D space from a single angled view
  • The position of the tracker and 3D image of the heart are co-registered and transferred into an MR image in real-time, visualized by see-through video glasses
  • PoC Data: Optimized machine learning models for locating the catheter have demonstrated a Euclidian distance error of <2 mm for certain test data sets

Technology Applications

  • Real-time, mixed reality visualization for cardiac interventional surgery
  • Improved preoperative planning for cardiac interventions

Technology Advantages

  • Catheter is visualized in 3D space in real time, allowing for better navigation
  • Models reflect individual patients’ heart architectures, enabling precise and individualized surgeries
  • Quantitative feedback will provide real-time guidance and post-intervention analytics

Figure of target 1 advance to SVC

Intellectual Property

Patents

  • US Application Filed

Cornell Reference

  • 9607

Contact Information

Donna Rounds, Ph.D

For additional information please contact

Donna Rounds
Associate Director, Business Development and Licensing
Phone: (646) 962-7044
Email: djr296@cornell.edu