Machine Learning Collateral Circulation Score Computation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional methods for assessing collateral coronary arteries are invasive, costly, time-consuming, and not suitable for routine clinical practice, lacking effective non-invasive and automated approaches to quantify collateral circulation for improved patient stratification and clinical decision-making.

Innovation Solution

A machine learning-based system that computes a collateral circulation score using patient data, including medical images, demographic, and lab/genetic information, employing trained networks for anatomical and functional assessments, and generating synthesized images to determine the functioning of collateral arteries, enabling automated and non-invasive evaluation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional invasive pressure and velocity measurements are used to assess collateral arteries, then measurement precision is improved, but device complexity and ease of operation deteriorate due to invasiveness and procedural complexity

Engineering Contradiction:
Improvecollateral artery assessment accuracyVSAvoidroutine clinical practice suitability
Core Design Contradiction:
Measurement precisionVSEase of operation

Solution Approach 1:

The patent replaces invasive mechanical pressure and velocity measurement systems with a non-invasive machine learning-based image analysis system. The system uses coronary angiography images as input and applies trained neural networks to automatically assess collateral artery function, eliminating the need for invasive catheter-based measurements while maintaining assessment accuracy.

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

Solution Approach 2:

The patent creates a virtual model of collateral artery function by training machine learning networks on synthesized and real angiography data. The trained model copies the assessment capabilities of invasive measurements by analyzing standard coronary angiography images, enabling non-invasive functional assessment that mirrors invasive measurement precision.

Inventive Principle:
Principle #26Copying

2Measurement precision

If manual annotations and observations are used for surrogate functional assessment, then measurement precision is improved, but productivity deteriorates due to time-consuming manual processes

Engineering Contradiction:
Improvesurrogate functional assessment accuracyVSAvoidassessment speed
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent implements a self-service assessment system where the machine learning model automatically performs collateral artery evaluation without requiring manual annotations or observer expertise. The trained network processes coronary angiography images and generates functional assessments autonomously, eliminating time-consuming manual processes while maintaining measurement precision.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent replaces manual observation and annotation processes with automated machine learning-based image analysis. The system substitutes human expert time with computational algorithms that rapidly process angiography images to provide functional assessments, significantly improving productivity while preserving assessment accuracy.

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

3Ease of operation

If anatomical assessment by measuring artery size and length is performed, then ease of operation is improved, but measurement precision deteriorates as clinical outcome correlation is not confirmed

Engineering Contradiction:
Improveanatomical measurement simplicityVSAvoidclinical outcome correlation
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The patent transforms the assessment from simple anatomical parameters (size, length) to functional parameters (flow index, pressure index, resistance index, velocity flow reserve) using machine learning analysis of angiography images. This parameter transformation maintains ease of operation while significantly improving clinical outcome correlation by assessing actual collateral function rather than just anatomical presence.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent replaces simple anatomical measurement techniques with advanced machine learning-based functional assessment. The system analyzes coronary angiography images to compute functional indices that correlate with clinical outcomes, substituting basic morphological evaluation with sophisticated computational analysis that provides clinically relevant functional information.

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

Data Source

PatentEP3819909A1Assessment of collateral coronary arteries
Publication Date: 2021.05.12 SIEMENS HEALTHINEERS AG
  • EP3819909A1 patent drawingFigure 1
  • EP3819909A1 patent drawingFigure 2
  • EP3819909A1 patent drawingFigure 3

AI summary

Systems and methods are provided for assessing collateral circulation of a patient. Patient data of a patient is received. A collateral circulation score is computed based on the patient data using a trained machine learning network. The collateral circulation score represents functioning of collateral circulation of the patient. The collateral circulation score is output.