Patient Surface Tracking via Weighted ICP Registration

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current medical data processing methods for tracking patient position during radiotherapy or radiosurgery face challenges in achieving real-time, accurate surface reconstruction, especially with non-coplanar treatment fields and spontaneous patient movement, as they often rely on high-speed approaches that compromise accuracy and are prone to local minima issues with iterative closest point (ICP) algorithms.

Innovation Solution

A method that uses a laser device to scan the patient's body surface, acquiring initial reflection data and subsequent scans, with positional information from later scans given lower weights to detect movement by variance comparison, allowing for continuous and efficient tracking even when parts of the body surface are outside the camera's field of view, and integrating range cameras into radiotherapy systems for stable surface registration with limited field of view.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If high-speed surface reconstruction approaches are used to track patient position in real-time, then tracking speed is improved, but measurement accuracy deteriorates

Engineering Contradiction:
Improvetracking speedVSAvoidsurface reconstruction accuracy
Core Design Contradiction:
SpeedVSMeasurement precision

Solution Approach 1:

The system performs preliminary surface scanning to acquire a reference surface model before treatment begins. This pre-acquired reference serves as a baseline for subsequent real-time tracking, allowing the system to compare current surface positions against the known reference geometry without requiring complex real-time reconstruction of the entire surface model.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The surface tracking problem is divided into two distinct phases: (1) offline reference surface acquisition using structured light scanning, and (2) online real-time tracking using range cameras. Each phase uses optimized methods appropriate to its requirements, with the reference phase prioritizing accuracy and the tracking phase prioritizing speed.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If iterative closest point (ICP) algorithm is used for surface registration, then alignment accuracy is improved, but computational complexity increases causing local minima issues

Engineering Contradiction:
Improvesurface registration accuracyVSAvoidalgorithm complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system performs preliminary surface scanning to acquire a reference surface model before treatment begins. This pre-acquired reference serves as a baseline for subsequent real-time tracking, allowing the system to compare current surface positions against the known reference geometry without requiring complex real-time reconstruction of the entire surface model.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The reference surface model serves multiple functions: it provides the baseline for ICP registration, acts as a geometric constraint to prevent convergence to local minima, and enables tracking even when parts of the patient body move outside the camera field of view.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Speed

If range cameras are used for real-time surface tracking, then tracking speed is improved, but measurement accuracy deteriorates due to limited field of view

Engineering Contradiction:
Improvereal-time tracking speedVSAvoidfield of view coverage
Core Design Contradiction:
SpeedVSArea of stationary object

Solution Approach 1:

The system creates a digital copy of the patient's surface geometry through structured light scanning before treatment. This virtual surface model can be viewed from any angle and position, compensating for the limited physical field of view of the range cameras during real-time tracking.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The system transitions from relying solely on direct optical observation (2D camera views) to incorporating a complete 3D surface model. This allows tracking information to be derived from the full surface geometry even when only portions are visible to the cameras at any given moment.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

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 enables efficient, continuous, and accurate tracking of patient position, reducing the risk of local minima and improving accuracy in non-coplanar positioning, allowing for markerless setup workflows and automatic surface-based prepositioning, thereby enhancing the reliability of radiotherapy systems.

Implementation Method 1

detection of light reflections from the surface of the body by scanning the body with a laser device and receiving the laser signals reflected from the body

Methodology Applied
Scientific EffectLight reflection: Reflection

Data Source

PatentEP3175426B1Online patient reconstruction and tracking for patient setup in radiation therapy using an iterative closest point algorithm
Publication Date: 2019.09.25 BRAINLAB AG
  • EP3175426B1 patent drawingFigure 1~3
  • EP3175426B1 patent drawingFigure 4
  • EP3175426B1 patent drawingFigure 5

AI summary

A medical data processing method for tracking the position of a body surface of a patient's body, the method comprising determining, based on initial surface reflection data and reflection pattern registration data, body surface movement data describing whether the body surface has undergone a movement.