Endoscope Multi-Wavelength Image Alignment for Oxygen Saturation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing endoscope systems face challenges in accurately aligning images taken with different wavelengths, which affects the precise calculation of oxygen saturation levels in blood vessels, especially for surface and subsurface vessels, due to variations in absorption and scattering characteristics of hemoglobin and digestive tract mucosa.
Innovation Solution
An endoscope system that applies first and second illumination lights with different wavelength ranges to a region of interest, using a color image sensor with pixels of at least three primary colors to acquire color signals, calculates shift amounts between frames based on same-color signals, and aligns images for precise registration and oxygen saturation level calculation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If images are acquired with different wavelength lights sequentially, then oxygen saturation levels can be calculated for different depth blood vessels, but positional shifts occur between frames causing misalignment
Solution Approach 1:
The patent creates a reference image from one wavelength frame and generates shifted versions of other wavelength frames by comparing them with the reference image. This copying approach allows the system to detect and correct positional shifts between frames acquired at different wavelengths, ensuring accurate alignment while maintaining the ability to detect both surface and subsurface blood vessels.
Solution Approach 2:
The patent adjusts the shift amount parameter based on comparisons between reference images and shifted images. By dynamically changing the shift amount to optimize alignment, the system achieves precise registration of multi-wavelength images while preserving the depth-specific detection capabilities of different wavelength ranges.
2Ease of operation
If shift amount is calculated between images of different wavelengths, then alignment can be attempted, but accurate registration is difficult due to varying absorption and scattering characteristics
Solution Approach 1:
Instead of directly comparing images of different wavelengths which have varying absorption and scattering characteristics, the patent creates a reference image and generates shifted copies of other wavelength frames. This copying method enables comparison of corresponding structures across wavelengths while accounting for wavelength-dependent image characteristics, improving registration accuracy.
Solution Approach 2:
The patent uses a reference image as an intermediary to facilitate alignment between different wavelength frames. By comparing all wavelength frames against a common reference rather than directly against each other, the system overcomes the difficulties of direct multi-wavelength comparison while maintaining operational simplicity.
3Measurement precision
If multiple wavelength lights are used to detect surface and subsurface blood vessels, then comprehensive oxygen saturation measurement is possible, but image alignment becomes more complex
Solution Approach 1:
The patent segments the image processing into distinct steps: creating a reference image from one wavelength, generating shifted versions of other wavelength frames, comparing them to detect shift amounts, and applying corrections. This segmentation of the complex multi-wavelength alignment process into manageable stages reduces overall system complexity while maintaining measurement precision.
Solution Approach 2:
The patent performs preliminary alignment by creating a reference image and generating shifted versions before final oxygen saturation calculation. This preliminary action of pre-aligning the frames simplifies subsequent processing steps and reduces the complexity of the overall system while ensuring accurate multi-wavelength measurement.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach allows for precise alignment and accurate calculation of oxygen saturation levels, enabling effective imaging of both surface and subsurface blood vessels, improving the detection of hypoxic regions such as cancerous tissues.
Implementation Method 1
a reflection image is imaged with a color image sensor
Implementation Method 2
an oxygen saturation level is calculated using the illumination light in a wavelength range in which absorbance varies depending on an amount of the hemoglobin in blood
Data Source
Figure 1
Figure 2
Figure 3
AI summary
First and second white light is generated by excitations of phosphors (50) with first and second laser beams having center wavelengths of 473 nm and 445 nm, respectively. The first and second white light is applied, in respective frames, sequentially to a region of interest in a subject. A color image sensor (60) images the region of interest in the each frame. Based on a shift amount, calculated from green signals of first and second frames, between images, an image of a blue signal of the first frame is moved to be aligned with an image of a green signal and an image of a red signal of the second frame. After the alignment, an oxygen saturation image representing an oxygen saturation level of hemoglobin in blood is produced from the blue signal of the first frame and green and red signals of the second frame, and displayed on a display (14).