Markerless Anatomy Tracking With Point Cloud Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing navigation systems in surgical procedures rely on marker-based tracking, which is invasive, time-consuming, and introduces latency, or markerless tracking methods suffer from high latency due to processing large data volumes and occlusions, especially when soft tissues obscure the view of bones.
Innovation Solution
A markerless tracking method using RGBD and structured light cameras generates high-density point clouds, segments anatomical features, and deforms mesh models to accurately track patient anatomy in real-time, employing preprocessing techniques like cropping, filtering, and edge processing to reduce computational load and latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If markerless tracking is used to avoid invasive procedures, then ease of operation is improved, but latency increases due to processing large data volumes
Solution Approach 1:
The patent segments the point cloud data into distinct anatomical regions (bone, soft tissue, air) using machine learning classification. This segmentation allows the system to process only relevant data portions for tracking, reducing overall computational load and latency while maintaining the non-invasive markerless approach.
Solution Approach 2:
The patent extracts and processes only the necessary anatomical features from the full point cloud data. By identifying and isolating key tracking points and anatomical landmarks, the system reduces the volume of data that requires intensive processing, thereby decreasing latency while preserving the benefits of markerless tracking.
2Measurement precision
If high-density point clouds are generated for accurate tracking, then measurement precision is improved, but processing time increases due to large data volumes
Solution Approach 1:
The patent applies different processing quality levels to different regions of the point cloud. High-density processing is applied only to critical anatomical regions requiring precise tracking, while less critical areas receive reduced processing. This local quality approach maintains measurement precision for essential tracking points while reducing overall processing time.
Solution Approach 2:
The patent processes only the essential portions of the point cloud data at full resolution, while using simplified processing for other regions. This partial action approach focuses computational resources on the most critical tracking elements, achieving sufficient measurement precision without the excessive processing time that would result from processing all data at maximum detail.
3Measurement precision
If soft tissue obstructions are accounted for to improve tracking accuracy, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent introduces machine learning-based segmentation as an intermediary process between raw point cloud capture and tracking analysis. This intermediary layer automatically distinguishes between soft tissue and bone structures, allowing the tracking system to focus on bone landmarks while filtering out soft tissue obstructions. This approach improves measurement precision without requiring complex hardware modifications.
Solution Approach 2:
The patent replaces complex mechanical or optical solutions for penetrating soft tissue with a computational approach. Instead of using specialized sensors or invasive methods to see through soft tissue, the system uses machine learning algorithms to interpret point cloud data and identify bone structures beneath soft tissue, reducing device complexity while improving tracking accuracy.
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
This approach reduces surgical setup time, minimizes invasive procedures, and achieves low-latency, real-time tracking of patient anatomy, even under soft tissue obstructions, by optimizing data processing and reducing reliance on preoperative imaging.
Implementation Method 1
generating a point cloud from image data obtained from one or more cameras... imaging, using the one or more cameras, the at least one object
Data Source
AI summary
A method for modeling patient anatomy includes, generating and segmenting a point cloud of the patient anatomy to identify visible bone surface and soft tissue surface regions of the patient anatomy, registering a reference bone model to the identified visible bone surface regions to provide an initial patient anatomy model having an estimation of a pose of bone portions of the patient anatomy, augmenting the initial patient anatomy model with soft tissue bodies based on the estimation of the pose of the bone portions, where the augmenting provides a full patient anatomy model having (i) the soft tissue bodies representing soft tissue portions of the patient anatomy and (ii) elements representing the bone portions of the patient anatomy, and registering the full patient anatomy model to the segmented point cloud such that the full patient anatomy model accurately reflects a current position and pose of the patient anatomy.