Pre-contrast Image Analysis for Dynamic Contrast Injection Parameter Optimization

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The challenge in medical imaging is achieving high-quality contrast images due to the interplay between acquisition, injection, and physiological parameters, making it difficult to adapt protocols dynamically and ensure personalized image quality.

Innovation Solution

A method using a convolutional neural network to process pre-contrast images, determining optimal injection parameters to achieve target image quality, and iteratively adjusting parameters based on real-time feedback from contrast images to ensure consistent image quality.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If acquisition parameters and injection parameters are fixed based on standard protocols, then the imaging process is simple and fast, but image quality varies over time and cannot be adapted to individual patient physiology

Engineering Contradiction:

Improveimage quality consistencyVSAvoidparameter adaptation system

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system performs preliminary analysis of pre-contrast images to predict optimal injection parameters before the actual contrast injection occurs. This allows the system to pre-determine the best acquisition and injection parameter combinations tailored to each patient's anatomy and physiology, ensuring consistent image quality without adding complexity during the actual imaging process

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system uses automated machine learning models to independently analyze pre-contrast images and self-determine optimal injection parameters without requiring manual intervention from technologists. The model automatically adapts parameters based on patient-specific features, making the system self-adjusting and eliminating the need for complex manual protocol adjustments

Inventive Principle:

Principle #25Self-service

2Reliability

If manual adjustment of acquisition and injection parameters is performed to adapt to physiological parameters, then personalized image quality can be achieved, but the process becomes time-consuming and complex

Engineering Contradiction:

Improvepersonalized image qualityVSAvoidparameter optimization time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system replaces manual mechanical adjustment of parameters by technologists with an automated machine learning-based parameter optimization system. The ML model processes pre-contrast images and automatically determines optimal injection and acquisition parameters, substituting human expertise and manual adjustment with an automated computational system that delivers personalized results instantly

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system dynamically changes injection parameters (such as contrast agent volume, injection rate, and timing) and acquisition parameters based on patient-specific features extracted from pre-contrast images. The machine learning model identifies the optimal parameter combinations for each patient, enabling personalized imaging protocols without manual intervention

Inventive Principle:

Principle #35Parameter changes

3Reliability

If real-time adaptation of parameters during injection sequence is implemented, then image quality can be maintained, but the system complexity and computational requirements increase significantly

Engineering Contradiction:

Improveimage quality during injection sequenceVSAvoidreal-time parameter adaptation system

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system performs all parameter optimization and adaptation calculations before the contrast injection begins, using pre-contrast images to predict the optimal parameter sequence for the entire injection protocol. This preliminary action eliminates the need for complex real-time parameter adjustments during the injection sequence, maintaining image quality without requiring continuous computational intervention

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system uses pre-contrast image analysis to predict patient-specific physiological responses and pre-determines the optimal injection and acquisition parameter sequence. This feedback mechanism allows the system to anticipate quality variations and pre-adjust parameters, avoiding the need for complex real-time monitoring and adjustment systems during the actual injection

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS20240054648A1Methods for training at least a prediction model, or for processing at least a pre-contrast image depicting a body part prior to an injection of contrast agent using said prediction model

Publication Date: 2024.02.15 GUERBET SA

AI summary

The present invention relates to a method for processing at least a pre-contrast image depicting a body part prior to an injection of contrast agent, the method being characterized in that it comprises the implementation, by a data processor (11b) of a second server (1b), of steps of: (a) Obtaining said pre-contrast image; (b) Determining candidate value(s) of at least one injection parameter of said injection of contrast agent by application of a prediction model to said pre-contrast image, such that a theoretical contrast image depicting said body part during injection of contrast agent in accordance with the determined candidate value(s) of said injection parameter(s) is expected to present a target quality level.