Digital microfluidics (DMF) is a versatile technique for parallel and field-programmable control of individual droplets. Given the high level of variability in droplet manipulation, it is essential to establish self-adaptive and intelligent control methods for DMF systems that are informed by the transient state of droplets and their interactions. However, most related studies focus on droplet localization and shape recognition. In this study, we develop the AI-assisted DMF framework μDropAI for multistate droplet control on the basis of droplet morphology. The semantic segmentation model is integrated into our custom-designed DMF system to recognize the droplet states and their interactions for feedback control with a state machine. The proposed model has strong flexibility and can recognize droplets of different colors and shapes with an error rate of less than 0.63%; it enables control of droplets without user intervention. The coefficient of variation (CV) of the volumes of split droplets can be limited to 2.74%, which is lower than the CV of traditional dispensed droplets, contributing to an improvement in the precision of volume control for droplet splitting. The proposed system inspires the development of semantic-driven DMF systems that can interface with multimodal large language models (MLLMs) for fully automatic control.

Digital microfluidic (DMF) has been a unique tool for manipulating micro-droplets with high flexibility and accuracy. To extend the application of DMF for automatic and in-site detection, it is promising to introduce colorimetric sensing based on gold nanoparticles (AuNPs), which have advantages including high sensitivity, label-free, biocompatibility, and easy surface modification. However, there is still a lack of studies for investigating the movement and stability of AuNPs for in-site detection on the electrowetting-based digital microfluidics. Herein, to demonstrate the ability of DMF for colorimetric sensing with AuNPs, we investigated the electrowetting property of the AuNPs droplets on the hydrophobic interface of the DMF chip and examined the stability of the AuNPs on DMF as well as the influence of evaporation to the colorimetric sensing. As a result, we found that the electrowetting of AuNPs fits to a modified Young–Lippmann equation, which suggests that a higher voltage is required to actuate AuNPs droplets compared with actuating water droplets. Moreover, the stability of AuNPs was maintained during the processing of electrowetting. We also proved that the evaporation of droplets has a limited influence on the detections that last several minutes. Finally, a model experiment for the detection of Hg²⁺ was carried out with similar results to the detections in bulk solution. The proposed method can be further extended to a wide range of AuNPs-based detection for label-free, automatic, and low-cost detection of small molecules, biomarkers, and metal ions.