Luminescence Image Segmentation for Clearer Tumor Resection Guidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Luminescence imaging in medical procedures is hindered by spurious light from foreign objects, which biases fluorescence image statistics, leading to misclassification of target bodies and incomplete resection during surgeries.
Innovation Solution
Process luminescence images limited to an informative region identified in an auxiliary image, using semantic segmentation to separate target bodies from foreign objects, and apply auto-scaling or thresholding to enhance tumor visibility.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If luminescence imaging is performed on the entire field of view, then complete coverage of the surgical area is achieved, but spurious light from foreign objects biases the fluorescence image statistics and reduces measurement precision
Solution Approach 1:
The patent segments the field of view into an informative region (surgical cavity containing target body) and a non-informative region (foreign objects) using semantic segmentation of auxiliary images. This allows processing to be focused on the informative region, eliminating spurious light contamination from foreign objects while maintaining complete coverage of the surgical area.
Solution Approach 2:
The patent extracts and processes only the informative region from the complete field of view. By identifying and isolating the surgical cavity in auxiliary images and applying the same spatial mask to luminescence images, the system extracts the relevant fluorescence signal while discarding contaminated regions containing foreign objects.
2Illumination intensity
If fluorescence values are converted to display range using the entire image statistics, then full dynamic range utilization is achieved, but the representation of the target body becomes less conspicuous due to spurious light bias
Solution Approach 1:
The patent applies segmentation by computing fluorescence statistics (minimum, maximum, mean values) exclusively from pixels within the informative region identified in auxiliary images. This segmented statistical approach ensures that display range conversion is based only on relevant tumor signal, maximizing tumor conspicuity while utilizing the full display dynamic range.
3Extent of automation
If thresholding is applied to the entire fluorescence image, then automatic tumor discrimination is achieved, but misclassification occurs due to biased statistical distribution from spurious light
Solution Approach 1:
The patent segments the image processing by calculating threshold values only from fluorescence values within the informative region. This segmentation ensures that automatic thresholding is performed on unbiased statistical data, eliminating misclassification errors caused by spurious light from foreign objects while maintaining automation.
Solution Approach 2:
The patent introduces auxiliary images as an intermediary step to identify the informative region. These auxiliary images serve as a mediator that guides the subsequent processing of luminescence images, enabling accurate region identification that improves the reliability of automatic thresholding.
4Measurement precision
If manual adjustments are made to limit spurious light effects, then measurement precision can be improved, but device complexity and operational difficulty increase
Solution Approach 1:
The patent implements self-service by using auxiliary images to automatically identify the informative region and generate spatial masks that are then applied to luminescence images. This self-identifying mechanism eliminates the need for manual region definition or operator intervention, maintaining measurement precision while avoiding increased device complexity.
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
Reduces false positives and negatives, ensuring complete resection of tumors by improving tumor visibility and reducing adverse effects of spurious light.
Implementation Method 1
Luminescence imaging is based on a luminescence phenomenon, consisting of the emission of light by luminescence substances when subject to any excitation different from heating
Implementation Method 2
particularly, a fluorescence phenomenon occurs in fluorescence substances (called fluorophores), which emit (fluorescence) light when they are illuminated
Implementation Method 3
an auxiliary image (based on an auxiliary light different from this luminescence light) of a field of view
Data Source
Figure 1
Figure 2A
Figure 2B
AI summary
A solution is proposed for assisting a medical procedure. A corresponding method comprises acquiring a luminescence image (205F), based on a luminescence light, and an auxiliary image (205R), based on an auxiliary light different from this luminescence light, of a field of view (103); the field of view (103) contains a region of interest comprising a target body of the medical procedure (containing a luminescence substance) and one or more foreign objects. An auxiliary informative region (210Ri) representative of the region of interest without the foreign objects is identified in the auxiliary image (205R) according to its content, and a luminescence informative region (210Fi) is identified in the luminescence image (205F) according to the auxiliary informative region (210Ri). The luminescence image (205F) is processed limited to the luminescence informative region (210Fi) for facilitating an identification of a representation of the target body therein. A computer program and a corresponding computer program product for implementing the method are also proposed. Moreover, a computing device for performing the method and an imaging system comprising it are proposed. A medical procedure based on the same solution is further proposed.