System and method for automated initialization and registration of navigation system
Inventors
Kadoury, Samuel • Kruecker, Jochen • Jago, James Robertson • Wood, Bradford Johns • Collet-Billon, Antoine • Dufour, Cecile
Assignees
Koninklijke Philips NV • US Department of Health and Human Services
Publication Number
US-10290076-B2
Publication Date
2019-05-14
Expiration Date
2032-03-02
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Abstract
A system and method for image registration includes tracking (508) a scanner probe in a position along a skin surface of a patient. Image planes corresponding to the position are acquired (510). A three-dimensional volume of a region of interest is reconstructed (512) from the image planes. A search of an image volume is initialized (514) to determine candidate images to register the image volume with the three-dimensional volume by employing pose information of the scanner probe during image plane acquisition, and physical constraints of a pose of the scanner probe. The image volume is registered (522) with the three-dimensional volume.
Core Innovation
The invention provides a system and method for image registration that tracks a scanner probe position along a patient's skin surface and acquires image planes from this position. From the image planes, a three-dimensional volume of a region of interest is reconstructed. A search of an image volume is then initialized to determine candidate images to register with the reconstructed three-dimensional volume. This search employs pose information of the scanner probe during image acquisition and physical constraints related to the probe's pose necessary for image acquisition. Finally, the image volume is registered with the three-dimensional volume.
The problem solved by the invention is that traditional fiducial marker-based spatial registration methods for aligning pre-operative high-resolution three-dimensional images, such as CT or MR scans, with intra-operative imaging modalities like ultrasound, are often infeasible or inaccurate. This infeasibility arises due to situations where fiducials could not be attached prior to imaging or when field of view limitations prevent simultaneous imaging of fiducials and regions of interest. Existing alternatives rely on manual registration that is time-consuming and prone to inaccuracy due to organ motion, deformation, and differing image modalities' characteristics.
The present invention addresses these issues by automatically initializing and optimizing image registration without reliance on manual alignment or fiducial markers. It uses tracked intra-operative images, integrated with image-based segmentation and shape information such as organ boundaries, to provide reliable registration between pre-operative volumes and intra-operative ultrasound images. This includes constraining the search space for registration based on patient positioning, probe pose, and physical constraints, enabling accurate transformation matrix estimation and subsequent optimization using similarity measures between images to align pre-operative and intra-operative image volumes.
Claims Coverage
The patent includes two independent claims covering a method and a system for image registration with specific innovative features.
Method for automated image registration using tracked scanner probe and physical pose constraints
The method involves tracking a scanner probe in at least one position along a patient’s skin surface, acquiring corresponding image planes via an imaging device, reconstructing a three-dimensional volume from those image planes, initializing a search to find candidate images to register the image volume with the three-dimensional volume by employing pose information of the scanner probe and necessary physical pose constraints related to the scanner probe, and registering the image volumes.
System for image registration integrating tracked scanner probe, imaging device, and processing modules with pose constraints
The system comprises a spatially trackable scanner probe, an imaging device configured to acquire image planes corresponding to at least one position of the scanner probe, an image processing module that reconstructs a three-dimensional volume and initializes a search for candidate images employing pose information and physical pose constraints of the scanner probe, and a registration module configured to register the image volume with the three-dimensional volume.
The claims collectively cover innovative features of tracking the scanner probe’s pose during acquisition, applying physical constraints for probe positioning, reconstructing 3D volumes from image planes, initializing constrained search spaces for registration, and registering the image volumes, encompassed in both method and system claims.
Stated Advantages
Provides rapid and accurate automatic or semi-automatic registration without reliance on fiducial markers.
Enables reliable subject-specific initialization for image registration that is independent of image quality.
Improves registration performance compared to existing fiducial marker-based methods.
Contributes to better multimodal registration between pre-operative and intra-operative images, facilitating procedures such as image-guided surgery and diagnostics.
Documented Applications
Verification and monitoring of fiducial-based registration between electromagnetic tracked ultrasound images and pre-acquired three-dimensional images.
Estimation of tumor or target locations on pre-operative images.
Improved initialization for image-based registration of two-dimensional ultrasound and pre-acquired three-dimensional images.
Non-rigid correction to registration of two-dimensional ultrasound and pre-acquired three-dimensional images.
Verification of tumor existence and assessment in diagnostic settings.
Guidance in interventional and surgical procedures including biopsy, ablation, embolization, and drainage.
Diagnostic applications employing multi-modality registration to ensure evaluation of the same lesion across different imaging modalities.
Interested in licensing this patent?