Endoscope Image Processing for Bleeding Stopping Point Guidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
During surgical procedures, accurately identifying the bleeding stopping point is challenging due to difficulties in recognizing blood vessel running, especially when blood vessels are deep or obscured by a pool of blood, which can lead to ineffective bleeding cessation.
Innovation Solution
An image processing device and method that acquire and process endoscope images, recognize blood vessel running information, and superimpose a bleeding stopping recommended point on the surgery image, considering the blood vessel's position and bifurcation, to guide the surgeon in performing hemostasis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If blood vessel running information is recognized from surgery images, then bleeding stopping point can be identified, but recognition accuracy deteriorates when blood vessels are deep or obscured by blood pools
Solution Approach 1:
The patent introduces a pool of blood virtual model as an intermediary element that mediates between the surgery image and the blood vessel running information. This virtual model represents the spatial distribution and optical properties of blood pools, enabling the system to compensate for obscured blood vessels by mathematically reconstructing the underlying vascular structure despite the interfering blood pool layer
Solution Approach 2:
The patent replaces direct visual mechanical observation of blood vessels with computational image processing and virtual model-based analysis. By substituting the mechanical/visual detection system with a computer-based virtual model system, the patent enables accurate blood vessel recognition even when visually obscured by blood pools
2Ease of operation
If bleeding stopping recommended point is superimposed on surgery image, then surgeon guidance is improved, but system complexity increases
Solution Approach 1:
The patent implements a self-service mechanism where the system automatically performs blood vessel recognition, pool of blood virtual model construction, and bleeding stopping point identification without requiring manual intervention from the surgeon. The superimposition of the recommended point on the surgery image is performed automatically, allowing the surgeon to simply follow the visual guidance without needing to understand or operate the complex underlying system
Data Source
AI summary
An image processing device includes a processor. The processor is configured to: recognize blood vessel running information; recognize a first bleeding position from a surgery image; identify a second bleeding position in the blood vessel running information corresponding to the first bleeding position in the surgery image; and identify a bleeding stopping point corresponding to the second bleeding position as a bleeding stopping recommended point corresponding to the first bleeding position. The processor: acquires information about the bleeding stopping recommended point corresponding to each of one or more blood vessel areas in the blood vessel running information; identifies the blood vessel area among the one or more blood vessel areas, to which the second bleeding position belongs; and performs processing in which the bleeding stopping recommended point corresponding to the blood vessel area thus identified is superimposed on the surgery image and displayed on a display.