Radiation Therapy Planning Engine for Real-Time Toxicity Prediction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The complexity of cancer treatment planning, particularly in radiation therapy, is hindered by the variability in patient-specific factors and the sporadic involvement of physicians and physicists, leading to challenges in predicting toxicity and treatment efficacy, which limits informed decision-making by clinical care teams.

Innovation Solution

A computer-implemented treatment planning engine that processes patient data, including images and medical history, to generate predicted treatment outcomes and matches, allowing clinicians to visualize and adjust treatment plans in real-time, thereby enabling informed tactical trade-offs during radiation treatment planning.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If radiation treatment planning is performed manually by dosimetrists with sporadic physician and physicist involvement, then flexibility in decision-making is maintained, but prediction accuracy of toxicity and treatment efficacy deteriorates

Engineering Contradiction:
Improveprediction accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The treatment planning system is segmented into distinct functional modules: a planning module that generates treatment plans, a prediction module that forecasts toxicity and efficacy outcomes, and an optimization module that refines plans based on predictions. This segmentation allows each module to specialize in specific tasks, improving overall prediction accuracy while maintaining manageable system complexity through modular architecture.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system introduces an intermediary prediction layer between treatment plan generation and clinical decision-making. This prediction module acts as a mediator that translates complex treatment parameters into actionable toxicity and efficacy forecasts, enabling clinicians to make informed decisions without requiring deep dosimetry expertise, thus improving prediction accuracy while preserving clinical flexibility.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If multiple treatment plans are evaluated with detailed toxicity and efficacy analysis, then treatment optimization is improved, but time consumption deteriorates

Engineering Contradiction:
Improvetreatment optimizationVSAvoidtime consumption
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary toxicity and efficacy predictions during the treatment planning stage itself, rather than requiring separate analysis phases. The prediction module generates outcome forecasts concurrently with plan evaluation, allowing clinicians to see predicted outcomes immediately alongside treatment parameters, thus improving treatment optimization without significant time penalty.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system replaces manual, time-consuming iterative evaluation of multiple treatment plans with an automated prediction and optimization engine. The computer-implemented prediction module rapidly evaluates toxicity and efficacy outcomes for multiple plans, substituting manual analysis with algorithmic processing that maintains high reliability while dramatically reducing time consumption.

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

3Quantity of substance

If homogeneous patient population data from clinical trials is used, then data availability is improved, but applicability to individual patients deteriorates

Engineering Contradiction:
Improvedata availabilityVSAvoidpatient-specific applicability
Core Design Contradiction:
Quantity of substanceVSAdaptability or versatility

Solution Approach 1:

The system applies local quality by tailoring the prediction model to individual patient characteristics. While training data comes from homogeneous clinical trial populations, the prediction module adjusts its forecasts based on patient-specific factors including anatomy, disease stage, and treatment parameters. This allows the system to leverage abundant trial data while adapting predictions to the unique characteristics of each individual patient.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system changes parameters by incorporating patient-specific variables into the prediction model. The prediction module dynamically adjusts toxicity and efficacy forecasts based on individual patient data such as age, comorbidities, tumor characteristics, and treatment plan specifics. This parameter adaptation transforms generalizable trial data into personalized predictions, improving patient-specific applicability while maintaining data availability from large trial cohorts.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3334496B1Result-driven radiation therapy treatment planning
Publication Date: 2021.06.30 SIRIS MEDICAL INC
  • EP3334496B1 patent drawingFigure 1
  • EP3334496B1 patent drawingFigure 2
  • EP3334496B1 patent drawingFigure 3

AI summary

The treatment planning engine empowers radiation treatment decision makers, such as a clinician, to efficiently identify optimal radiation treatment results for a given patient. Specifically, the treatment planning engine generates a set of recommended treatment results based on patient images, patient contours, and prior patient data. A user interface simultaneously displays the patient images and contours with the recommended treatment results. The clinician is able to adjust the contours in the user interface and, in real-time, evaluate the impact on the tumor volume and the toxicity risk to the nearby anatomical structures. With the user interface, the clinician can visualize the relationship between radiation delivery to the contours and the resulting recommended treatment results, allowing the clinician to make informed decisions about certain trade-offs during radiation treatment planning for the patient. The treatment delivery engine generates a radiation treatment plan based on a recommended treatment result selected by the clinician.