Blood Cell Classification via Flow Imaging and ML
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cell counting methods in blood samples, such as flow cytometry, lack morphological information and require manual preparation of slides, which is time-consuming and inefficient.
Innovation Solution
A method and system that captures frames of a blood sample as it flows, segments and tracks objects, classifies them using image processing and machine learning techniques, and computes overall classification results to improve accuracy without manual slide preparation, incorporating morphological data for more precise cell counting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If flow cytometry is used to count cells, then cell counting speed is improved, but morphological information is lost
Solution Approach 1:
The patent combines flow cytometry technology with digital imaging technology to create a hybrid system. The flow cell maintains the rapid single-file cell passage of flow cytometry, while integrated cameras capture morphological images of each cell. This merging allows simultaneous acquisition of both quantitative counting data and qualitative morphological data without sacrificing either aspect.
Solution Approach 2:
The patent transitions from traditional two-dimensional flow cytometry measurements to multi-dimensional analysis by incorporating digital images that capture cellular morphology, shape, and structural features. This adds a visual dimension to the quantitative data, enabling classification based on both numerical parameters and morphological characteristics.
2Loss of information
If manual slide preparation is used for morphological analysis, then cell morphology is preserved, but time consumption increases
Solution Approach 1:
The patent replaces the manual mechanical process of slide preparation, staining, and microscopy with an automated digital imaging system. Cells pass through a flow cell where cameras automatically capture images, eliminating the need for manual slide handling while preserving morphological integrity through controlled flow conditions and optical imaging.
Solution Approach 2:
The system performs self-service by automatically capturing, storing, and analyzing cell images as cells flow through the device. The digital images are immediately processed and archived, eliminating the need for external manual intervention in slide preparation and morphological assessment.
3Productivity
If automated cell counting is implemented, then productivity is improved, but accuracy of cell classification deteriorates
Solution Approach 1:
The system incorporates feedback mechanisms where captured images are analyzed by classification algorithms that compare cellular features against reference databases. The system can iteratively refine classifications based on image quality and cellular characteristics, improving accuracy while maintaining automated operation. Operator feedback can also be integrated to correct misclassifications.
Solution Approach 2:
The patent uses composite classification approaches that combine multiple data sources including flow cytometry parameters, image processing features, and machine learning algorithms. This multi-parameter composite analysis improves classification accuracy by leveraging the strengths of different methodologies simultaneously rather than relying on a single automated technique.
Data Source
AI summary
Systems and methods for classifying blood cells in a blood sample are disclosed. A series of frames of the blood sample as it flows through a field of view of an image capture device are captured and analysed. Advantageously, the disclosed systems and methods combine the availability of morphological cell data with the convenience of a flow-through arrangement. The classification results can be used for estimating cell counts in a blood sample.