Blood Vessel Segmentation Using Multi-Channel Black Blood MRI

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current medical imaging techniques for cerebrovascular diseases are inefficient and inaccurate, particularly in the automated assessment of vessel segmentation and detection of vascular abnormalities, which hinders timely and precise diagnosis and treatment.

Innovation Solution

A system and method utilizing multiple imaging modalities, including black blood MRI, to generate and process images using a blood vessel segmentation model and registration techniques, enabling automated detection of vascular structures and abnormalities.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional medical imaging techniques are used for cerebrovascular disease diagnosis, then the imaging process can be performed with conventional equipment, but the processing efficiency and accuracy are insufficient

Engineering Contradiction:
Improveimage processing efficiencyVSAvoidvascular detection accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent replaces traditional manual image processing methods with an automated machine learning-based system. A blood vessel segmentation model processes medical images to automatically identify and segment blood vessels, replacing manual analysis with computational algorithms that achieve both high efficiency and accuracy in vascular detection

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

Solution Approach 2:

The patent transforms image processing from conventional methods to deep learning-based methods by changing the fundamental parameters of the processing system. The blood vessel segmentation model uses neural network parameters to extract vascular information, enabling simultaneous improvement in processing speed and detection precision

Inventive Principle:
Principle #35Parameter changes

2Loss of information

If single-modality imaging is used, then the imaging process is simple and quick, but comprehensive vascular information including both walls and lumens cannot be obtained

Engineering Contradiction:
Improvevascular information completenessVSAvoidimaging system complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The patent merges multiple imaging modalities (black blood images showing vascular walls and bright blood images showing vascular lumens) into a unified processing framework. The blood vessel segmentation model integrates information from both modalities to produce comprehensive vascular segmentation results, eliminating information loss while managing system complexity through integrated processing

Inventive Principle:
Principle #5Merging (Combining)

3Productivity

If manual blood vessel segmentation is performed, then the process is flexible and adaptable, but the processing time is long and productivity is low

Engineering Contradiction:
Improveimage processing throughputVSAvoidoperational simplicity
Core Design Contradiction:
ProductivityVSEase of operation

Solution Approach 1:

The blood vessel segmentation model performs self-service by automatically processing and segmenting blood vessels in medical images without requiring manual intervention. The trained neural network independently analyzes images, identifies vascular structures, and generates segmentation results, dramatically increasing processing throughput while maintaining operational simplicity through automated execution

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP4579593B1Systems and methods for image processing
Publication Date: 2026.05.20 SHANGHAI UNITED IMAGING HEALTHCARE
  • EP4579593B1 patent drawingFigure 1
  • EP4579593B1 patent drawingFigure 2
  • EP4579593B1 patent drawingFigure 3

AI summary

The present disclosure provides methods and systems for image processing. The methods may include obtaining multiple black blood images of a first region of a target subject. The methods may further include generating a first segment result by inputting the black blood images into multiple input channels of a blood vessel segmentation model. The first segment result may include blood vessel segments of first blood vessels in the first region, the blood vessel segmentation model may be a trained machine learning model, and each of the black blood images may be randomly designated as an input image of one of the input channels.