Object Geometry Determination Using ML Segmentation and Hybrid Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing navigation systems struggle to accurately and efficiently determine the geometry of objects, such as implants, relative to anatomical structures during procedures like spinal fusion, due to limitations in image segmentation and tracking technologies.
Innovation Solution
A system and method utilizing machine learning algorithms, particularly convolutional neural networks, to automatically segment and identify anatomical features in image data, combined with electromagnetic and optical tracking to navigate and display the pose of instruments relative to the patient, enabling precise implant positioning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If machine learning algorithms are used to automatically segment and identify anatomical features, then measurement precision and productivity are improved, but device complexity increases
Solution Approach 1:
The system employs machine learning algorithms that automatically segment and identify anatomical features from image data without requiring manual intervention or complex user input. The Convolutional Neural Network processes images autonomously to extract geometric information, reducing the need for manual measurement and analysis while improving precision.
Solution Approach 2:
Traditional manual measurement and geometric determination methods are replaced with automated machine learning-based image processing. The system substitutes mechanical/manual operations with computational algorithms that can process medical images to extract anatomical geometries with high precision.
2Measurement precision
If electromagnetic and optical tracking systems are combined to navigate instruments, then measurement precision and reliability are improved, but device complexity increases
Solution Approach 1:
The system merges electromagnetic tracking and optical tracking technologies to create a hybrid navigation system. By combining the strengths of both tracking methods, the system achieves improved measurement precision and reliability in determining instrument positions relative to anatomical structures during procedures.
Solution Approach 2:
The tracking system is designed to perform multiple functions using both electromagnetic and optical modalities. This multi-functional approach allows the system to track instruments in various configurations and environments, enhancing the system's versatility and reliability in surgical navigation.
3Productivity
If real-time processing of image data and tracking information is implemented, then productivity and ease of operation are improved, but use of energy increases
Solution Approach 1:
The system continuously processes image data and tracking information in real-time throughout the procedure, maintaining continuous updates of anatomical geometries and instrument positions. This continuous processing enables real-time navigation decisions and improves workflow efficiency by eliminating delays between data acquisition and analysis.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables efficient and accurate determination of object geometry and pose, facilitating precise procedural planning and execution by superimposing tracked instruments onto segmented anatomical images in real-time, enhancing surgical accuracy and workflow efficiency.
Implementation Method 1
measuring an effect of an electromagnetic (EM) field on a sensor coil
Implementation Method 2
a current is induced in the sensor coil. The measured induced current may be used to identify or determine a position of the instrument
Implementation Method 3
determining a location with optical sensors
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
Disclosed is a system to assist in a procedure. During the procedure an object may be moved relative to a subject, such as being positioned and/or placed within a subject. The system and related method may be used to assist in displaying and/or determining a pose of the object relative to a subject, such as rigid portions of a subject.