Image-Guided Tissue Isolation for Spatial Biochemical Molecules
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for isolating biochemical molecules in spatial omics lose spatial context and require additional preparation time, potentially damaging cells, or are inefficient in target detection and isolation.
Innovation Solution
A method and system utilizing image processing and neural networks to automatically detect and isolate biochemical molecules by imaging tissue samples, extracting spatial information, and physically isolating targets using lasers or other methods based on this information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If FACS or MACS are used to sort cells, then cell isolation can be performed, but spatial context is lost before sorting
Solution Approach 1:
The tissue sample is divided into multiple sections mounted on separate glass slides, allowing spatial context to be preserved in each section while enabling parallel processing. The system segments the tissue architecture into manageable units that maintain their spatial relationships during analysis and isolation.
Solution Approach 2:
The system creates digital copies (images) of the tissue sections that serve as references for automated target detection and isolation. These digital representations allow the system to locate and isolate targets without physically disrupting the spatial context of the original tissue architecture.
2Loss of information
If LCM is used to sort ROIs while maintaining spatial context, then spatial context is preserved, but additional preparation time is required and cells may be damaged
Solution Approach 1:
The system performs preliminary imaging of the tissue sections to create digital references before the isolation process. This allows automated detection algorithms to be pre-trained on the spatial patterns, enabling rapid target identification without requiring time-consuming manual preparation or UV laser focusing during the isolation phase.
Solution Approach 2:
The system replaces manual mechanical operations (UV laser focusing, manual dissection) with automated image processing and computer-controlled isolation mechanisms. This substitution eliminates the need for operator skill in focusing UV beams and reduces cell damage from mechanical dissection or IR-activated polymer melting.
3Ease of operation
If conventional cell sorters are used, then cell sorting can be performed, but detection and isolation are inefficient
Solution Approach 1:
The system enables self-service automated detection where the digital images of tissue sections automatically provide the spatial information needed for target identification. The imaging system and detection algorithms work autonomously to locate targets without requiring manual intervention or complex manual sorting procedures, significantly improving detection and isolation speed.
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
Enables rapid and precise isolation of biochemical molecules while preserving spatial context, reducing detection and isolation time.
Implementation Method 1
imaging a tissue sample to acquire a tissue image in the form of any one of a multi-channel fluorescence image and a visible light image
Implementation Method 2
physically isolating and retrieving the target from the tissue sample
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present invention relates to a method and system for isolating biochemical molecules, and the method for isolating biochemical molecules may include the steps of: imaging a tissue sample to acquire a tissue image in the form of any one of a multi-channel fluorescence image and a visible light image; obtaining and analyzing detection results for each channel from the tissue image to extract spatial information of a target, or analyzing the tissue image through a learned neural network model to predict spatial information of a target; and adjusting the size and location of a isolation area in a biochemical molecule isolation device based on the spatial information of the target, and then physically isolating and retrieving the target from the tissue sample.