Knowledge-Based Brachytherapy Planning Using Primitive Features

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current brachytherapy treatment planning for prostate cancer is resource-intensive and requires expert skill, leading to a decline in utilization due to high technical difficulty and inadequate training, with existing automated systems unable to replicate plans created by experienced practitioners.

Innovation Solution

A computer-implemented method using a knowledge-based algorithm that generates brachytherapy plans by computing primitive features from patient contour data, querying a database for similar plans, and refining them using a stochastic search algorithm to create clinically feasible plans in near real-time.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If manual planning is performed by expert radiation oncologists, then plan quality is high, but resource consumption and time requirements are excessive

Engineering Contradiction:

Improveplan qualityVSAvoidresource consumption

Core Design Contradiction:

Manufacturing precisionVSProductivity

Solution Approach 1:

The system creates a database of expert-generated treatment plans that serve as templates. New treatment plans are generated by copying and adapting these proven expert plans rather than creating them from scratch, thereby preserving high plan quality while reducing the time and expertise required for each new case.

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The system pre-processes and stores multiple expert treatment plans in a database with standardized formats and parameters. This preliminary preparation allows the automated system to quickly retrieve and adapt appropriate templates during treatment planning, eliminating the need for extensive manual work each time.

Inventive Principle:

Principle #10Preliminary action

2Manufacturing precision

If manual planning is performed by expert radiation oncologists, then plan quality is high, but the process is resource intensive and requires expert skill

Engineering Contradiction:

Improveplan qualityVSAvoidtechnical difficulty

Core Design Contradiction:

Manufacturing precisionVSDevice complexity

Solution Approach 1:

The system enables automated treatment planning that does not require expert intervention for each case. The automated algorithm independently retrieves templates, adapts them to patient-specific anatomy, and generates treatment plans, making the complex expert knowledge self-contained within the software rather than requiring expert operators.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system replaces the manual mechanical process of expert planning with an automated computational algorithm. The algorithm uses image processing, template matching, and optimization routines to automatically generate treatment plans, substituting human expert manual work with automated computational mechanisms.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Productivity

If automated treatment planning options are used, then resource consumption is reduced, but they cannot mimic plans created by expert planners

Engineering Contradiction:

Improveresource efficiencyVSAvoidplan quality

Core Design Contradiction:

ProductivityVSManufacturing precision

Solution Approach 1:

The system directly copies proven expert treatment plans from the database as templates. By basing automated plans on actual expert-generated cases rather than theoretical algorithms, the system preserves the quality characteristics of expert planning while maintaining automated efficiency.

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The system uses a database of expert plans as an intermediary between expert knowledge and automated generation. Rather than trying to encode expert knowledge directly into algorithms, the system mediates through stored example plans that bridge the gap between human expertise and automated processing.

Inventive Principle:

Principle #24Intermediary (Mediator)

4Manufacturing precision

If expert skill is required for brachytherapy planning, then plan quality is maintained, but training requirements and staffing needs increase

Engineering Contradiction:

Improveplan qualityVSAvoidtraining requirements

Core Design Contradiction:

Manufacturing precisionVSEase of operation

Solution Approach 1:

The system makes expert-level plan generation accessible without requiring operators to become experts. The automated algorithm performs the complex planning tasks independently, allowing staff with minimal training to generate high-quality treatment plans that previously required hundreds of hours of expert training.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS10967199B2Knowledge-based brachytherapy planning based on primitive features computed from anatomical contour data

Publication Date: 2021.04.06 SUNNYBROOK RES INST

AI summary

Described here are systems and methods for knowledge-based brachytherapy planning. These systems and methods are capable of automatically generating treatment plans for prostate brachytherapy in clinically relevant times. Primitive features computed from anatomical contours of the target organ are used to retrieve a template plan from a database. The template plan is then adjusted in a stochastic search algorithm.