Radiation Inverse Treatment Planning via Sparse Dose Shot Selection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current radiation inverse treatment planning systems for linear accelerators are not sufficiently precise, require a lengthy planning process, and are not capable of real-time adaptation to patient movements, leading to inadequate protection of surrounding areas and increased treatment sessions.

Innovation Solution

A radiation inverse treatment planning system that pre-computes a set of individual dose shots with predetermined locations, angles, sizes, and shapes, and associates weights to each shot based on constraints, using a sparse linear combination of beams from a pre-defined dictionary to perform real-time calculations and adapt to patient movements.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If traditional radiation treatment planning systems are used, then treatment can be delivered, but the planning process is lengthy and not sufficiently precise

Engineering Contradiction:
Improvedose distribution precisionVSAvoidplanning process duration
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The system pre-computes a comprehensive dictionary of potential dose shots covering all possible locations, incidence angles, sizes, and shapes before treatment planning begins. This preliminary preparation allows the actual planning to quickly select from pre-calculated options rather than computing each shot from scratch, thereby achieving high precision dose distribution without lengthy planning processes.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The treatment planning is segmented into two distinct phases: (1) pre-computation of a comprehensive dictionary of dose shots with all possible parameters, and (2) interactive selection and weighting of specific shots from this dictionary. This segmentation allows the computationally intensive work to be done beforehand, enabling fast and precise treatment planning when needed.

Inventive Principle:
Principle #1Segmentation

2Adaptability or versatility

If traditional treatment planning systems are used, then treatment planning can be completed, but real-time adaptation to patient movements is not possible

Engineering Contradiction:
Improvereal-time adaptation capabilityVSAvoidprotection of surrounding areas
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The system transforms static treatment planning into a dynamic process by maintaining a pre-computed dictionary of dose shots that can be rapidly re-weighted and re-selected in real-time based on patient movement. This allows the treatment plan to adapt dynamically to changing conditions while ensuring reliable protection of surrounding areas through continuous optimization.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system incorporates real-time feedback mechanisms where patient position and movement are continuously monitored, and the treatment plan is automatically adjusted by re-weighting shots from the pre-computed dictionary. This feedback loop ensures that the actual dose delivery matches the planned distribution even when patient movements occur, maintaining reliability of surrounding area protection.

Inventive Principle:
Principle #23Feedback

3Manufacturing precision

If multiple successive incidences are used to increase conformity, then dose delivery precision is improved, but the number of treatment shots and planning complexity increase

Engineering Contradiction:
Improvedose conformityVSAvoidnumber of treatment shots
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The pre-computed dictionary serves as a universal repository containing dose shots with all possible locations, angles, sizes, and shapes. This single comprehensive dictionary can address all treatment requirements and optimization scenarios, eliminating the need for separate planning processes for different shot configurations and reducing overall system complexity despite using multiple successive incidences.

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

Solution Approach 2:

The system achieves dose conformity by varying parameters of pre-computed shots (weighting factors, selection of different shots from the dictionary) rather than by increasing the fundamental complexity of the treatment delivery system. This allows high conformity through parameter optimization of a manageable number of shots rather than requiring an excessive number of incidences.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS9700738B2System and computer program product for radiation inverse treatment planning
Publication Date: 2017.07.11 INTUITIVE THERAPEUTICS SA
  • US9700738B2 patent drawing
  • US9700738B2 patent drawing
  • US9700738B2 patent drawing

AI summary

The present invention concerns a radiation inverse treatment planning system for a linear accelerator. The system includes a radiation source, configured for delivering individual radiotherapeutic dose shots (aj), each individual radiotherapeutic dose shot having a predetermined location and incidence angle inside and/or outside a target area, a size and a shape. The system also includes at least a data bus system (102), and a memory (106) coupled to the data bus system (102), wherein the memory (106) includes a computer usable program code. The system also includes a processing unit (104) coupled to the data bus system (102), wherein the processing unit (104) is configured to execute the computer usable program code to pre-compute (10) a set of individual radiotherapeutie dose shots (aj), and associate (40) a weight (sj) to each individual radiotherapeutic dose shot (aj), based on one or more constraints (20). The processing unit (104) executes the computer usable program code to find (30) the sparsest subset of individual radiotherapeutic dose shots, so as to satisfy said one or more constraints (20).