Monocular Endoscope Instrument Tracking for Surgical Distance Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing surgical systems using monocular cameras are unable to accurately estimate positional data during minimally-invasive surgeries, necessitating the use of cumbersome and inaccurate physical rulers for measurements.
Innovation Solution
A surgical system that estimates positional data using monocular cameras by matching a 3D model of a surgical instrument to images captured by the endoscope, adjusting for focal length and other parameters to determine the distance between two locations based on relative transformations of the instrument's pose.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a monocular camera is used to capture surgical images, then the device complexity is reduced and ease of operation is improved, but the ability to estimate positional data (distance measurement precision) deteriorates
Solution Approach 1:
The patent introduces an intermediary computational process that matches 3D models of surgical instruments to 2D image data from the monocular camera. By using 3D model matching algorithms, the system recovers depth information and positional data that would otherwise be unavailable from a single 2D image, thereby enabling distance estimation without requiring complex stereoscopic hardware
Solution Approach 2:
The patent replaces the mechanical/optical solution of using stereoscopic cameras (which require complex hardware configuration) with a computational approach. The system substitutes physical depth-capturing mechanisms with image processing algorithms that analyze 2D images and reconstruct 3D positional information through computer vision techniques
2Measurement precision
If physical rulers are used for measurements during surgery, then measurement capability is provided, but ease of operation deteriorates due to cumbersome handling and accuracy reduces
Solution Approach 1:
The system enables self-service measurement capability by automatically calculating distances between surgical sites directly from images captured during the procedure. The computational system processes the image data and provides measurement results without requiring the surgeon to manually introduce or handle physical measuring tools, making the measurement process as convenient as image capture itself
3Measurement precision
If stereoscopic cameras are used to enable distance estimation, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent creates a computational 3D model (a virtual copy) of the surgical scene based on 2D image data. By matching known 3D models of surgical instruments to the captured images and using pose recovery algorithms, the system reconstructs three-dimensional positional information from two-dimensional inputs, effectively creating a virtual stereoscopic representation without physical stereoscopic cameras
Data Source
AI summary
A method performed by a surgical system. The method includes receiving a first image from an endoscope, the first image having a surgical instrument at a first location within a field of view of the endoscope. The method matches a three-dimensional (3D) model of the surgical instrument to the first image. The method receives a second image receives a second image from the endoscope, the second image having the surgical instrument at a second location within the field of view of the endoscope, and matches the 3D model of the surgical instrument to the second image. The method estimates a distance between the first and second locations based on both matchings and one or more parameters of the endoscope.