3D Motion Estimation from 2D MRI Slices for Adaptive Radiotherapy

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current MRI-guided radiotherapy systems face challenges in accurately localizing and tracking targets and organs at risk in 3D space, especially due to significant out-of-plane motion and the difficulty in simultaneously tracking both targets and organs at risk, which limits effective dose calculations and adaptive radiotherapy.

Innovation Solution

A computer-implemented method and system that estimate 3D motion from a series of 2D MRI slices by building a conversion model in a learning stage and applying it in a tracking stage to provide real-time 3D motion estimation, enabling accurate tracking of target position, deformation, and rotation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If 2D MRI slices are used for target localization, then imaging speed and soft tissue contrast are improved, but 3D motion localization accuracy deteriorates due to out-of-plane motion

Engineering Contradiction:
Improveimaging speedVSAvoid3D motion localization accuracy
Core Design Contradiction:
SpeedVSMeasurement precision

Solution Approach 1:

The patent applies dimensionality change by transforming 2D MRI slices into 3D motion information through a conversion model. The system takes 2D motion field estimation from slices and converts it to 3D motion field representation, enabling accurate 3D target localization while maintaining the speed advantages of 2D imaging. This resolves the contradiction by adding the third dimension through computational transformation rather than requiring actual 3D imaging.

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

2Ease of operation

If sequential 2D slices are acquired alternating axial, coronal and sagittal, then target motion tracking is enabled, but simultaneous tracking of organs at risk becomes difficult

Engineering Contradiction:
Improvetarget motion tracking capabilityVSAvoidorgan at risk tracking
Core Design Contradiction:
Ease of operationVSDifficulty of detecting and measuring

Solution Approach 1:

The patent applies universality by creating a conversion model that serves multiple functions simultaneously. The same 2D slice data and conversion model can track both targets and organs at risk in 3D space, eliminating the need for separate tracking approaches. The model universally handles different anatomical structures and motion types from the same 2D imaging data.

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

3Productivity

If only 2D slice information is gathered during treatment, then real-time tracking is possible, but offline retrospective dose calculation becomes difficult

Engineering Contradiction:
Improvereal-time tracking efficiencyVSAvoid3D anatomical information for dose calculation
Core Design Contradiction:
ProductivityVSLoss of information

Solution Approach 1:

The patent applies preliminary action by pre-building a conversion model during a learning stage that transforms 2D slice data into 3D motion field representations. This pre-computed model enables both real-time tracking and comprehensive offline dose calculations to be performed on the same 2D slice data, eliminating the need to choose between speed and information completeness.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3721941B1Motion management in MRI-guided linac
Publication Date: 2025.03.05 ELEKTA AB
  • EP3721941B1 patent drawingFigure 1
  • EP3721941B1 patent drawingFigure 1B
  • EP3721941B1 patent drawingFigure 2

AI summary

Described herein is a system and method of controlling real-time image-guided adaptive radiation treatment of at least a portion of a region of a patient. The computer-implemented method comprises obtaining a plurality of real-time image data corresponding to 2-dimensional (2D) magnetic resonance imaging (MRI) images including at least a portion of the region, performing 2D motion field estimation on the plurality of image data, approximating a 3-dimensional (3D) motion field estimation, including applying a conversion model to the 2D motion field estimation, determining at least one real-time change of at least a portion of the region based on the approximated 3D motion field estimation, and controlling the treatment of at least a portion of the region using the determined at least one change.