Visually Intelligent Digital Microfluidics for Generalized Colorimetric Assays

The application of digital microfluidics in biochemical reaction analysis has received widespread attention and is becoming more mature, but the degree of automation in reaction and result reading still needs to be improved, especially when a large number of droplets are present. Therefore, this paper realizes the detection and visual intelligent control for large-scale droplet arrays based on the artificial intelligence object detection model, and proposes a mask method for the readout of the colorimetric reaction results, predicts the reaction results using deep learning models, and carries out the application research in a variety of colorimetric reactions using active-matrix digital microfluidics.