Image Processing Apparatus for Subsurface Blood Vessel Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In surgical and endoscopic procedures, it is challenging for doctors to visually recognize blood vessels located beneath the biological surface, leading to increased procedural difficulty and reliance on anatomical knowledge or visual cues.
Innovation Solution
An image processing apparatus that acquires video data, estimates the spatial distribution of biological components with temporal variations, and generates periodic variation video data by applying a filter to extract specific frequency components, thereby enhancing the visibility of subtle vascular movements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a doctor performs surgery by visually recognizing blood vessels on the biological surface, then the procedure is straightforward when vessels are visible, but the doctor cannot detect blood vessels located in layers below the surface, leading to increased procedural difficulty and reliance on anatomical knowledge
Solution Approach 1:
The patent applies color mapping to represent different tissue depths and blood vessel characteristics in the generated image. By assigning specific colors to different depths (e.g., red for superficial, blue for deep) and blood vessel types, the system enables doctors to visually distinguish between vessels at various layers, directly addressing the limitation of not being able to detect subsurface vessels
Solution Approach 2:
The system introduces an intermediate processing layer that includes a neural network model and image generation module. This intermediary processes the captured image data, extracts depth and vessel information, and generates an enhanced image that reveals subsurface structures, thereby bridging the gap between surface-level visual detection and deep vessel detection
2Measurement precision
If conventional video processing methods are used to capture surgical footage, then the recording is simple, but subtle variations such as blood vessel beats and pulse waves cannot be visually recognized
Solution Approach 1:
The system dynamically processes video frames by applying the neural network model to each frame or sequence of frames, extracting temporal variations in blood vessel characteristics. This dynamic processing enables detection of beating patterns and pulse waves by analyzing changes across multiple frames, thereby visualizing subtle temporal variations that static or simple processing would miss
Solution Approach 2:
The system changes key processing parameters including the neural network architecture (e.g., U-Net, ResNet), filtering parameters for extracting periodic components, and color mapping parameters. By adjusting these parameters, the system optimizes the extraction of subtle variations while managing computational complexity, enabling visualization of blood vessel dynamics without excessive processing burden
Data Source
AI summary
An image processing apparatus includes a processor including hardware. The processor is configured to: acquire video data; estimate a spatial distribution of a predetermined biological component for which a component spectrum temporally varies with respect to image data of a predetermined frame among a plurality of frames included in the video data; generate biological component video data in which the biological component is extracted from the video data based on the estimated spatial distribution of the biological component in the image data of the predetermined frame; and generate periodic variation video data of the biological component by applying, to the biological component video data, a filter for extracting a predetermined frequency component.