Microfluidic Flow Control Using CNN-Based Abnormality Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Microfluidic systems require time-intensive human monitoring and intervention to prevent disruptions due to abnormalities in fluid flow, which are subjective and time-consuming, leading to production losses and system damage.
Innovation Solution
A machine learning-based microfluidic control system that uses a convolutional neural network to classify images of the microfluidic device operation as normal or abnormal, automatically adjusting parameters such as flow rates to maintain optimal operation, including transitioning between regimes like dripping, jetting, breaking, and wetting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If human monitoring and intervention is used to prevent disruptions in microfluidic systems, then operational reliability is improved, but time consumption and subjectivity increase leading to production losses
Solution Approach 1:
The system performs self-monitoring and self-adjustment through automated image classification and flow rate control. The microfluidic system automatically detects abnormalities in droplet formation and adjusts fluid flow rates without human intervention, enabling the system to serve itself in maintaining optimal operation.
Solution Approach 2:
The patent replaces manual human monitoring and adjustment with an automated computational system. Image classification algorithms and automated control mechanisms substitute for human visual inspection and manual parameter adjustment, eliminating subjectivity and time consumption associated with human operation.
2Ease of operation
If human monitoring is used to assess fluid flow abnormalities, then operational control is achieved, but subjectivity and time consumption increase
Solution Approach 1:
The system automatically classifies images of droplet formation and identifies abnormalities without human assistance. The automated image analysis and abnormality detection enable the system to self-assess its operational state and trigger appropriate control actions.
Solution Approach 2:
The system implements continuous feedback loops where images of droplet formation are captured, classified, and used to automatically adjust fluid flow rates. This closed-loop feedback mechanism replaces human assessment with automated real-time monitoring and adjustment.
3Reliability
If human intervention is used to adjust flow rates and return to normal operation, then system recovery is achieved, but production loss increases due to subjective and time-consuming processes
Solution Approach 1:
The system automatically detects when droplet formation becomes abnormal and self-corrects by adjusting fluid flow rates through automated control mechanisms. This self-service capability enables rapid system recovery without human intervention.
Solution Approach 2:
The automated monitoring and control system operates continuously without interruption, ensuring that droplet formation remains optimal at all times. This continuous operation eliminates production losses that would occur during manual monitoring and adjustment intervals.
4Device complexity
If manual monitoring is used for complex parallelized arrays of microfluidic platforms, then operational oversight is provided, but the complexity makes effective monitoring and control difficult
Solution Approach 1:
Each microfluidic platform in the parallelized array performs self-monitoring and self-adjustment through automated image classification and flow rate control. This distributed self-service approach simplifies the operation of complex systems by eliminating the need for centralized manual monitoring.
Solution Approach 2:
The automated image classification and control system provides universal monitoring capability across all parallelized microfluidic platforms. A single automated system can simultaneously monitor and control multiple platforms, making the operation of complex arrays as easy as simple systems.
Data Source
AI summary
A system is provided to automatically monitor and control the operation of a microfluidic device using machine learning technology. The system receives images of a channel of a microfluidic device collected by a camera during operation of the microfluidic device. Upon receiving an image, the system applies a classifier to the image to classify the operation of the microfluidic device as normal, in which no adjustment to the operation is needed, or as abnormal, in which an adjustment to the operation is needed. When an image is classified as normal, the system may make no adjustment to the microfluidic device. If, however, an image is classified as abnormal, the system may output an indication that the operation is abnormal, output an indication of a needed adjustment, or control the microfluidic device to make the needed adjustment.