Microchannel Particle Sorting with Machine Learning Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional particle sorting techniques using microchannels are limited by their inability to accommodate varying viscosities and particle size distributions in biological samples, leading to inconsistent sorting accuracy and potential clogging issues, making them inconvenient for use with biological samples with great individual variation.
Innovation Solution
A particle sorting apparatus and method that employs a microchannel device controlled by a trained model generated through machine learning, using a computation unit to determine optimal control conditions based on control and separation data, allowing for precise adjustment of flow rate and viscosity to effectively sort particles by size.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional particle sorting technique using microchannel is applied to liquids with various viscosities, then particle sorting can be performed, but sorting accuracy varies and clogging may occur
Solution Approach 1:
The invention changes the control parameters (flow rate, viscosity) based on the sample characteristics. The system dynamically adjusts these parameters to accommodate different viscosities while maintaining consistent sorting accuracy, resolving the contradiction between versatility and reliability.
Solution Approach 2:
The invention implements a feedback mechanism where the system learns from separation results and adjusts control conditions accordingly. This feedback loop enables the system to adapt to varying viscosities and maintain reliable sorting performance across different sample types.
2Stability of the object's composition
If anticoagulants are added to reduce viscosity variation, then viscosity becomes more constant, but viscosity may become too high causing clogging
Solution Approach 1:
The invention dynamically adjusts the flow rate parameter based on the viscosity level. When viscosity is high (even with anticoagulants), the system increases flow rate to prevent clogging, thus maintaining both viscosity consistency and preventing harmful clogging effects.
Solution Approach 2:
The invention makes the flow rate dynamic rather than fixed. The system continuously adapts the flow rate according to the actual viscosity conditions, enabling it to handle both low and high viscosity samples without clogging while maintaining stable composition.
3Reliability
If device structure is optimized for specific viscosity, then sorting accuracy improves, but device complexity and production cost increase
Solution Approach 1:
The invention replaces complex mechanical device optimization with a computational approach. Instead of designing different physical devices for different viscosities, the system uses machine learning to determine optimal control parameters, substituting mechanical complexity with software-based adaptation.
Solution Approach 2:
The invention makes a single microchannel device universal by using learned control conditions that adapt to different sample types. The same device structure can handle various viscosities and particle sizes through parameter adjustment, eliminating the need for multiple specialized devices.
4Reliability
If flow rate is optimized for each sample viscosity, then sorting accuracy improves, but time and cost for optimization increase
Solution Approach 1:
The invention performs preliminary learning during a training phase where the system collects data and learns optimal control conditions for different sample types. This preliminary action stores the optimization results, so that subsequent sorting operations can directly apply learned conditions without time-consuming real-time optimization.
Solution Approach 2:
The invention uses feedback from separation results to continuously improve control conditions. The system learns from past separations and refines its parameter selection, reducing the time needed for optimization over time while maintaining high sorting accuracy.
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
This approach enables efficient and accurate particle sorting across varying viscosities and size distributions, improving sorting accuracy and reducing the risk of clogging, thus facilitating the use of the technique in biological samples with complex characteristics.
Implementation Method 1
The separation is achieved by utilizing a laminar flow that occurs at a point where two (bifurcated) channels merge, and based on the difference in forces applied to the flowing particles depending on the sizes of the particles.
Data Source
AI summary
A particle sorting apparatus for separating particles according to the sizes of the particles, and includes a microchannel device, a computation unit that determines a condition for controlling the microchannel device using a trained model obtained through machine learning of control condition data and separation result data that have been obtained by separating particles while controlling the microchannel device, and a control unit that controls the microchannel device based on the condition.