PET Image Processing for Automated SUV Normalization Parameters

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current PET image processing methods lack automation in calculating filter parameters for SUV normalization, leading to inefficiencies and inaccuracies due to manual calculations and inconsistent standard application across devices and patients, which affects SUV comparability and clinical utility.

Innovation Solution

An image processing system and method that calculates filter parameters based on phantom image data, automating the normalization process by matching acquisition and reconstruction parameters with a configuration file, allowing for flexible selection of SUV normalization standards like EARL V1.0 and V2.0, and providing intuitive graphical interfaces for user-friendly operation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If manual calculation of filter parameters is used for SUV normalization, then physicians can control the normalization process, but the process becomes time-consuming and lacks user-friendliness

Engineering Contradiction:

Improveuser-friendlinessVSAvoidprocessing time

Core Design Contradiction:

Ease of operationVSLoss of time

Solution Approach 1:

The system automatically retrieves filter parameters from a configuration file based on matching acquisition and reconstruction parameters, eliminating the need for manual calculation by physicians. The system serves itself by autonomously selecting appropriate parameters without human intervention, thereby reducing processing time while maintaining normalization quality

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

Filter parameters are pre-calculated and stored in a configuration file during system setup or previous operations. When processing PET images, the system simply retrieves these pre-prepared parameters by matching device and protocol identifiers, avoiding time-consuming manual calculations during actual clinical workflow

Inventive Principle:

Principle #10Preliminary action

2Productivity

If SUV normalization is performed manually with input boxes for filter parameters, then physicians can adjust parameters, but the method lacks efficiency and consistency across devices

Engineering Contradiction:

Improveprocessing efficiencyVSAvoidconsistency of standard application

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The configuration file stores filter parameters that are universally applicable across multiple PET devices and reconstruction protocols. By matching device identifiers and protocol codes, the system retrieves standardized parameters that ensure consistent SUV normalization across different devices, patients, and time points, eliminating variability introduced by manual parameter selection

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system establishes a feedback mechanism where acquisition parameters and reconstruction parameters from the PET device are automatically matched with stored configuration data. This closed-loop approach ensures that the correct filter parameters are selected based on actual device and protocol characteristics, maintaining both efficiency and reliability

Inventive Principle:

Principle #23Feedback

3Ease of operation

If filter parameters are manually entered at the workstation, then physicians can customize parameters for specific cases, but the process is not time-efficient

Engineering Contradiction:

Improveease of parameter inputVSAvoidnormalization speed

Core Design Contradiction:

Ease of operationVSProductivity

Solution Approach 1:

The manual mechanical process of physicians entering parameters through input boxes is replaced with an automated electronic retrieval system. The system electronically matches device identifiers, acquisition parameters, and reconstruction parameters with stored configuration data to automatically retrieve the appropriate filter parameters, dramatically increasing normalization speed while maintaining ease of use through automated workflows

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP4343778B1Image processing method, apparatus and system, and storage medium

Publication Date: 2026.02.18 SHANGHAI UNITED IMAGING HEALTHCARE

Figure 1
Figure 2
Figure 3

AI summary

The present disclosure provides an image processing method, a system, an apparatus, and a storage medium thereof. The method comprises obtaining a phantom image and a normalization standard (310); performing a first processing on the phantom image; generating a recovery coefficient of the processed phantom image generated by the first processing; determining whether the recovery coefficient satisfies the normalization standard (320); in response to determining that the recovery coefficient satisfies the normalization standard, obtaining a target processing parameter (330), wherein the target processing parameter is a processing parameter of the first processing when the recovery coefficient satisfies the normalization standard; and determining an image processing result by processing an image to be processed based on the target processing parameter (340).