Blood Cell Identification via Plate Compression and Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for identifying and differentiating white blood cells in blood samples are inefficient, as they often require complex equipment and procedures, and struggle to accurately distinguish between various cell types without causing cell deformation or damage.
Innovation Solution
A device comprising two movable plates with spacers that compress a blood sample into a thin layer, allowing for imaging and analysis using bright-field and fluorescence imaging, with a machine learning model to classify cell types based on image data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If complex equipment and procedures are used for identifying and differentiating white blood cells, then measurement precision may improve, but device complexity increases and ease of operation decreases
Solution Approach 1:
The device segments the sample analysis process into distinct functional zones within a single chip: a compression zone with parallel plates for cell layering, a staining zone for chemical treatment, and an imaging zone for capture. This segmentation allows complex functions to be distributed across simple, integrated components rather than requiring a single complex instrument.
Solution Approach 2:
The patent introduces an intermediary machine learning model that acts as a bridge between simple image capture and complex cell identification. The model processes images taken by a standard camera, automatically differentiating cell types based on learned patterns from training data, thereby eliminating the need for complex optical or mechanical identification systems.
2Measurement precision
If complex equipment and procedures are used for identifying and differentiating white blood cells, then measurement precision may improve, but ease of operation worsens
Solution Approach 1:
The device is designed to be largely self-operating: the sample automatically wicks through the chip via capillary action, the machine learning model automatically processes images and identifies cells, and results are generated without requiring user interpretation. This self-service design makes the device as easy to operate as applying a adhesive bandage, while maintaining high measurement precision through automated analysis.
3Measurement precision
If spacing between plates is reduced to compress sample into thin layer, then imaging quality improves, but cell deformation increases
Solution Approach 1:
The patent optimizes the spacing parameter between parallel plates to a specific range (2-10 micrometers) that balances two competing requirements: it is small enough to compress cells into a thin monolayer for high-quality imaging, yet large enough to avoid excessive deformation that would compromise cell integrity. This precise parameter control resolves the contradiction between imaging quality and cell shape preservation.
4Measurement precision
If multiple imaging modes (bright-field, dark-field, fluorescence) are used, then measurement precision improves, but device complexity and use of energy increase
Solution Approach 1:
The patent combines multiple imaging capabilities into a single fluorescence imaging step. By incorporating fluorescent stains that differentially label various cell structures and using a single fluorescence camera, the system achieves the discriminatory power of multiple imaging modes while using only one imaging modality, thereby reducing energy consumption and simplifying the device.
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 accurate and efficient identification and differentiation of white blood cells, including neutrophils, lymphocytes, monocytes, eosinophils, and basophils, without causing significant cell deformation, using a user-friendly device and machine learning-based analysis.
Implementation Method 1
reducing the spacing of the two plates to a dimension that is less than the dimension of the cell that is not compressed by the plates
Implementation Method 2
one of the images is a fluorescence image
Implementation Method 3
the stain uses a fluorescence dye and a surfactant together
Implementation Method 4
one of the images is a bright-field image taken by a white light
Data Source
AI summary
The disclosure provides a method for identifying a bio-entity including a cell type and count in a sample. The method includes: providing a device comprising a first plate, a second plate, and a patterned structural element; depositing the sample between the first and second plates; reducing the spacing of the first and second plates so that the first and second plates are in a closed configuration to compress the sample into a layer; and imaging the sample to obtain an image; and measuring and analyzing the image against a database generated with a machine learning model to obtain the bio-entity of the sample. The sample can be a blood sample, and the method can be a white blood cell differential test conducted with a mobile phone.