A novel method for texture-mapping conoscopic surfaces for minimally invasive image-guided kidney surgery

Area of Science:

• Medical Imaging
• Surgical Navigation
• Robotics

Background:

• Organ-level registration is vital for precise image-guided therapy, particularly for mobile soft tissues like the kidney.
• Accurate localization of organs is a persistent challenge in minimally invasive procedures.

Purpose of the Study:

• To develop and validate a novel registration method for soft tissue organs.
• To combine holographic conoscope 3D points with endoscopic surface data for improved tissue identification.
• To achieve accurate image-to-physical space registration for image-guided interventions.

Main Methods:

• A holographic conoscope and endoscopic camera were used to acquire 3D point clouds and textured surfaces.
• The conoscope's laser dot was localized in endoscopic space to register the textured surface to the point cloud.
• Standard surface registration techniques were applied after filtering points to the kidney surface.
• Spatial accuracy and temporal effects were evaluated using phantoms, ex vivo tissues, and an industrial robot.

Main Results:

• Surface acquisition time was approximately 55 seconds per hand-held acquisition.
• No significant temporal effects were observed related to acquisition synchronization or probe speed.
• The method achieved submillimeter target registration errors in applied target localization.

Conclusions:

• Textured surfaces were reconstructed with submillimetric mean registration errors.
• The developed method shows promise for image-guided therapy in various anatomical structures accessible via surgical techniques.
• This approach enhances precision in organ registration for improved therapeutic outcomes.