Radiological Image Vessel Identification via Zone Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In operational radiology, accurately identifying and targeting specific vessels in a complex vascular network for minimally invasive surgical procedures is challenging due to the difficulty in distinguishing between numerous vessels on radiological images, which limits precise treatment of pathologies without affecting adjacent segments.
Innovation Solution
A method for processing radiological images that allows operators to define a zone of interest and determine the corresponding part of the flow circulation network, highlighting the relevant vessels or zones with increased intensity or contrasting colors, aiding in the selection of optimal operation points and guiding tools for targeted treatment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional radiological imaging is used to guide catheter navigation, then the procedure remains minimally invasive, but the operator cannot accurately identify and target specific vessels in the complex vascular network
Solution Approach 1:
The patent segments the complex vascular network by allowing operators to define zones of interest and automatically identifying the corresponding vascular territories. The system divides the entire vascular network into manageable segments based on anatomical regions, enabling focused analysis of specific vessel groups rather than attempting to interpret the complete network at once.
Solution Approach 2:
The patent introduces an intermediary software system that acts as a mediator between the raw radiological images and the operator's decision-making. This intermediary automatically performs complex image processing, vessel segmentation, and territory identification, translating difficult-to-interpret imaging data into clear visual representations of vascular territories and their corresponding organ zones.
2Productivity
If operators manually identify vessels on radiological images, then the system remains simple, but the time required to identify all supplying vessels increases significantly
Solution Approach 1:
The patent implements preliminary action by automatically performing vessel identification and territory mapping before the operator needs to make treatment decisions. The system pre-processes the radiological images, identifies all vessels supplying the zone of interest, and prepares visual representations in advance, eliminating the time-consuming manual search process during the procedure.
Solution Approach 2:
The patent enables self-service by allowing the software system to automatically perform tasks that would otherwise require manual operator effort. The system autonomously identifies vessels, determines vascular territories, and generates visual displays without requiring the operator to manually trace or identify each vessel, significantly reducing preparation time.
3Loss of information
If conventional imaging displays all vessels in the network, then complete anatomical information is provided, but the relevant vessels for the target zone become difficult to distinguish
Solution Approach 1:
The patent applies local quality by providing different visual representations for different regions of the image. Zones supplied by vessels of interest are highlighted with enhanced visual properties (such as color coding or intensity changes), while other regions maintain their standard appearance. This allows the relevant vascular information to stand out without losing the context of the complete anatomical structure.
Solution Approach 2:
The patent adds another dimension to the visualization by overlaying vascular territory information onto the anatomical images. Instead of displaying vessels as simple linear structures, the system creates two-dimensional or three-dimensional representations of vascular territories that show the spatial distribution of blood supply, making it easier to distinguish relevant vessels from the surrounding network.
Data Source
AI summary
Embodiments of the invention relate to a method of processing a radiological image of an organ, the organ being connected to a flow circulation network, and comprising steps according to which:an operator or a processing means defines a contour to delimit a zone of interest in the radiological image of the organ;the processing means determines part of the network in which the flow supplies the zone of interest or originates from this zone of interest; anda display means displays the determined part of the network.