Circulating Abnormal Cell Detection Using Labeled Nuclei Signals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for detecting circulating abnormal cells (CACs) in peripheral blood rely on morphological information and manual intervention, leading to subjective and unreliable detection with low efficiency and high costs.
Innovation Solution
A method and device utilizing image processing algorithms and morphological algorithms to segment and label cell nuclei, followed by inputting labeled images into a pre-built circulating abnormal cell detection model to quantify staining signals and determine cell type based on acquired counts, leveraging a deep learning network for improved accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If conventional staining methods (May-Grunwald-Giemsa, Wright-Giemsa, etc.) are used for cell detection, then the staining process is simple and quick, but the staining results are easily affected by environmental factors and operator experience, leading to poor consistency and reliability
Solution Approach 1:
The patent replaces the manual mechanical staining process with an automated liquid dispensing system that uses robotic arms and precision pumps to deliver standardized amounts of staining reagents. This substitution eliminates operator variability and environmental influences, ensuring consistent staining results while maintaining process simplicity through automation.
Solution Approach 2:
The patent implements precise control of staining parameters including reagent volume (5-10 μL), incubation time (15-30 minutes), and temperature (20-25°C). By standardizing these parameters across all samples, the system achieves reproducible staining results independent of operator experience or environmental fluctuations.
2Device complexity
If manual cell counting and classification is performed, then the equipment required is simple, but the detection efficiency is low and productivity is limited
Solution Approach 1:
The patent replaces manual visual inspection and counting with an automated imaging system using high-resolution cameras and computer vision algorithms. The system automatically captures images of stained cells, processes them through machine learning models for classification, and generates quantitative results, thereby dramatically increasing productivity while maintaining relatively simple equipment architecture.
Solution Approach 2:
The system implements automated image processing and cell classification through pre-trained machine learning models that autonomously identify and categorize different cell types without requiring manual intervention. The entire workflow from sample preparation to result generation is self-executing, significantly enhancing detection efficiency.
3Reliability
If automated liquid handling systems are implemented to improve staining consistency, then the reliability of staining results improves, but the device complexity and cost increase
Solution Approach 1:
The automated system is divided into modular functional units: a liquid dispensing module for reagent delivery, an incubation module for controlled staining, an imaging module for cell capture, and an analysis module for data processing. Each module operates independently with standardized interfaces, allowing the system to achieve high reliability through modular automation while keeping individual component complexity manageable.
Data Source
Figure 1~2
Figure 3
Figure 4~5
AI summary
A method and device for detecting circulating abnormal cells. The method for detecting the circulating abnormal cells comprises: respectively segmenting and labelling, by using an image processing algorithm and a morphological algorithm, cell nuclei included in dark field microscope images of a plurality of probe channels (101); inputting the dark field microscope images, in which cell nuclei are labelled, of the plurality of probe channels into a pre-built circulating abnormal cell detection model to acquire the number of staining signals included in each labelled cell nucleus in the dark field microscope image of each probe channel (102); and for each labelled cell nucleus, on the basis of the number of the staining signals included in the labelled cell nucleus in the acquired dark field microscope image of each probe channel, determining whether the labelled cell nucleus belongs to a circulating abnormal cell (103). The method can effectively improve the reliability of detecting the circulating abnormal cells.