Quantifying capillary dynamics in paper-based microchannels using BS and openCV motion Algorithms

Paper-based microfluidics provides a low-cost platform for chemical and biological analyses, where understanding fluid flow dynamics is essential for ensuring accuracy and reliability. Traditional microfluidic paper-based analytical devices (μPADs) rely primarily on capillary action and imbibition to drive fluid through the porous structure of paper. These forces play a central role in defining the flow behavior within the microchannels. Accurate determination of flow parameters—such as velocity and acceleration—is critical for quantitative analysis, reaction efficiency, and reproducibility. This is particularly important in point-of-care (POC) diagnostic applications for medical or environmental testing, where precise timing and detection reliability are key. Flow parameters in μPADs are typically assessed by tracking the fluid meniscus position using a digital camera and extracting its position, z(t) , over the time, either manually or through automated scripts. The accuracy of this dynamic profile depends on both camera resolution and the extraction algorithm, as these affect how faithfully the imbibition behavior and fluid motion —driven by fluid properties and paper fiber structure—are captured. We developed a single-channel μPAD to correlate fluidic dynamics with the viscosity and composition of test fluids. The flow kinetics were extracted from video recordings using a high-performance software tool based on background subtraction (BS) algorithms and the Open Computer Vision library (OpenCV). The high accuracy in meniscus tracking enabled estimation of fluid composition from microfluidic velocity and viscosity. Importantly, the resulting flow curves revealed the presence of multiple dynamic regimes, strongly linked to the structural heterogeneity of cellulose fibers. • Developed a single-channel paper-based microfluidic device (μPAD) • Correlation of flow dynamics with fluid viscosity and composition. • Employed high-accuracy meniscus tracking with background subtraction OpenCV algorithms, enabling precise estimation of fluid composition from microfluidic velocity and viscosity. • Identified multiple dynamic flow regimes linked to cellulose fiber structural heterogeneity. • Method supports enhanced reliability in point-of-care (POC) diagnostic applications.