Markerless Camera Calibration for Image-Guided Surgery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image-guided surgery systems face challenges in accurately correlating high-quality 3D diagnostic images with distorted fluoroscopic images, leading to complex and time-consuming initialization protocols, and introducing inaccuracies due to the use of reference units on patients, which affects the precision of tool positioning and navigation.
Innovation Solution
A system and method for on-line calibration of imaging cameras using sensors positioned with respect to the imager and camera, characterizing transformations between coordinate systems to quantify intrinsic camera parameters without the need for fiducials or markers, allowing for dynamic tracking and correction of imaging system parameters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If reference units or markers are used to correlate 3D diagnostic images with fluoroscopic images, then the correlation accuracy is improved, but the complexity of the initialization protocol increases and measurement precision deteriorates due to introduced inaccuracies
Solution Approach 1:
The patent extracts and eliminates the markers/fiducials from the imaging system by using a markerless approach. Instead of relying on external reference units attached to patients or instruments, the system uses natural anatomical landmarks and image-based feature matching to establish correlations between 3D diagnostic images and fluoroscopic images, thereby reducing complexity while maintaining precision
Solution Approach 2:
The patent replaces the mechanical/physical marker-based registration system with a computational image processing system. The initialization protocol transitions from physically placing and tracking markers to using automated image feature detection, template matching, and computational geometry algorithms to correlate images, thereby reducing physical complexity while improving measurement precision
2Measurement precision
If markers or fiducials are used for coordinate registration, then the positioning accuracy is improved, but the time required for calibration increases
Solution Approach 1:
The patent performs preliminary actions by pre-processing and storing key anatomical landmarks and feature points from 3D diagnostic images before the actual surgical procedure. During calibration, the system rapidly matches these pre-prepared features with fluoroscopic images using efficient algorithms, eliminating the time-consuming process of manual marker placement and tracking while maintaining high positioning accuracy
Solution Approach 2:
The system uses self-service by automatically detecting and matching anatomical features without requiring external markers. The calibration process becomes self-contained, using the patient's own anatomy as the reference framework, which eliminates the time required for marker application, tracking, and removal while preserving positioning precision through automated image feature correlation
3Measurement precision
If markers are placed on patients for calibration, then the coordinate system correlation is improved, but the distortion in fluoroscopic images increases
Solution Approach 1:
The patent extracts and removes markers from the imaging field by using markerless image correlation. The system correlates coordinate systems by matching natural anatomical landmarks and geometric features directly visible in fluoroscopic images, eliminating the harmful artifacts and distortions that markers introduce while maintaining accurate coordinate transformation through computational geometry
Solution Approach 2:
The patent converts the challenge of image distortion into a benefit by using the distortion characteristics themselves as calibration information. The system models and characterizes the fluoroscopic image distortion mathematically, then uses this modeled distortion to improve coordinate correlation accuracy, turning what was previously a harmful artifact into a useful calibration parameter
Data Source
AI summary
Certain embodiments of the present invention provide methods and systems for calibration of an imaging camera or other image acquisition device. Certain embodiments include characterizing a transformation from a coordinate system of an imager to a coordinate system of a first sensor positioned with respect to the imager using a first off-line calibration. Certain embodiments also include characterizing a transformation from a coordinate system of an imaging camera source to a coordinate system of a second sensor positioned with respect to the imaging camera source using a second off-line calibration. Additionally, certain embodiments include quantifying intrinsic parameters of the imaging camera source based on a transformation from the coordinate system of the imager to the coordinate system of the imaging camera source based on the first and second off-line calibrations and information from the first and second sensors and a transmitter positioned with respect to an object being imaged.