Markerless 3D Lung Tumor Tracking via Anatomical Landmarks

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for tracking lung tumour motion during radiotherapy are inadequate, as they rely on invasive fiducial markers, suffer from limited tumour visibility due to anatomical obstructions, and provide only two-dimensional imaging, leading to inaccurate beam targeting and significant underdose in lung proton therapy.

Innovation Solution

A method for three-dimensional tracking of a target within a body using a processing system that processes two-dimensional scanned images to predict, measure, and estimate the target's position through a state transition model and statistical inference, enabling accurate tracking without markers and overcoming visibility limitations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If fiducial markers are implanted for tumour tracking, then tracking reliability is improved, but patient invasiveness and procedural complexity increase

Engineering Contradiction:
Improvetracking reliabilityVSAvoidpatient invasiveness
Core Design Contradiction:
ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The patent extracts the tracking function from invasive markers and relocates it to non-invasive anatomical landmarks and imaging data. The system uses visible anatomical structures (spine, ribs, pelvis) as reference points instead of implanted markers, eliminating the need for surgical intervention while maintaining tracking capability through image processing and coordinate transformation algorithms.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent introduces an intermediary computational system that bridges the gap between 2D imaging data and 3D tumour position. This system uses coordinate transformation, image registration, and statistical inference to derive accurate 3D tumour coordinates from 2D fluoroscopic images and 3D CT datasets, eliminating the need for direct physical markers on the tumour.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Device complexity

If 2D X-ray imaging is used for tumour tracking, then device complexity is reduced, but measurement precision deteriorates due to loss of depth information

Engineering Contradiction:
Improveimaging system complexityVSAvoidtracking precision
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent applies dimensionality change by combining multiple 2D fluoroscopic images taken at different gantry angles with a 3D CT dataset. Through coordinate transformation and image registration, the system reconstructs 3D tumour position from 2D projections, effectively adding the missing depth dimension without requiring complex 3D imaging during treatment.

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

Solution Approach 2:

The patent replaces the mechanical limitation of 2D imaging with computational methods. Instead of using complex 3D imaging hardware during treatment, the system uses mathematical algorithms (coordinate transformation, least-squares optimization, statistical inference) to extract 3D position information from 2D images, substituting mechanical complexity with computational processing.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Ease of operation

If markerless tracking is implemented, then patient comfort and procedure simplicity are improved, but tumour visibility and tracking reliability worsen due to anatomical obstructions

Engineering Contradiction:
Improveprocedure simplicityVSAvoidtracking reliability
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The patent makes the tracking system universal by using anatomical landmarks that are always visible regardless of tumour location or gantry angle. The system can track tumours in any position within the treatment field by referencing fixed anatomical structures (spine, ribs, pelvis) that serve multiple functions as both positioning references and visibility guarantees.

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

Solution Approach 2:

The patent performs preliminary action by pre-processing and registering the 3D CT dataset with the fluoroscopic imaging system before treatment begins. This pre-registration establishes coordinate transformations and identifies anatomical landmarks in advance, enabling reliable markerless tracking without requiring tumour visibility during the actual treatment procedure.

Inventive Principle:
Principle #10Preliminary action

4Measurement precision

If multiple gantry angles are used for imaging, then tracking accuracy is improved, but radiation exposure and treatment time increase

Engineering Contradiction:
Improvetracking accuracyVSAvoidtreatment time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent applies partial action by using a limited set of gantry angles (typically 0° and 180° for AP/PA views) rather than comprehensive multi-angle imaging. This partial imaging approach provides sufficient information for 3D position reconstruction through the pre-established coordinate transformations, avoiding the time and radiation burden of excessive angular sampling while maintaining adequate tracking accuracy.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3710109B1Three-dimensional tracking of a target in a body
Publication Date: 2023.08.23 LEO CANCER CARE INC
  • EP3710109B1 patent drawingFigure 1
  • EP3710109B1 patent drawingFigure 2
  • EP3710109B1 patent drawingFigure 3

AI summary

Disclosed is a method and system for three-dimensional tracking of a target located within a body, the method performed using at least one processing system. A two-dimensional scanned image of the body including the target is processed to obtain a two-dimensional image of the target. A first present dataset of the target is predicted using a previous dataset of the target and a state transition model, the first present dataset includes a three- dimensional present position value of the target. A second present dataset of the target is measured by template-matching of the two-dimensional image of the target with a model of the target. A third present dataset of the target is estimated by statistical inference using the first present dataset and the second present dataset. The previous dataset of the target is updated to match the third present dataset.