bioRxiv Science⌕ Search

Biology subjects

Lashkaripour, A.

Publications and source records attributed to Lashkaripour, A..

2 recordsLinked to original sources

Large field of view fluorescence imaging of microfluidic devices with a tandem-lens macroscope

Microfluidic devices enable high-throughput sample processing with remarkable parallelization and miniaturization. While fluorescence microscopy provides a convenient method for reading out signal from microfluidic assays, commercially-available microscopes impose a fundamental tradeoff between temporal resolution, spatial resolution, and numerical aperture (NA). Spatially tiled imaging enables high-resolution and high-NA imaging over a large area but reduces temporal resolution. Conversely, low magnification, low NA imaging captures large areas in one shot, but typically sacrifices spatial resolution and fluorescence sensitivity. To address this, we introduce an automated transfluorescence tandem-macro-lens optomechanical system (macroscope) capable of sensitive, multi-channel fluorescence imaging over a very large field of view (34 mm diameter, 740 mm2), with resolution determined by the sensor pixel size. We demonstrate bright-field resolution of low-micron features and detection of low-to mid-nanomolar concentrations of common fluorophores within microfluidic device channels. To demonstrate the utility of this macroscope, we image enzyme turnover within valved microfluidic devices (the HT-MEK system, for High-Throughput Microfluidic Enzyme Kinetics) and achieve >50-fold increased temporal resolution over common commercial instruments while maintaining high sensitivity. This macroscope imaging solution costs substantially less than commercially available alternatives, providing a powerful new imaging approach for microfluidic applications requiring sensitive and rapid wide-field fluorescence imaging.

bioengineering↗

Design Automation of Microfluidic Single and Double Emulsion Droplets with Machine Learning

Droplet microfluidics enables kHz screening of picoliter samples at a fraction of the cost of other high-throughput approaches. However, generating stable droplets with desired characteristics typically requires labor-intensive empirical optimization of device designs and flow conditions that limit adoption to specialist labs. Here, we compile the most comprehensive droplet dataset to date and use it to train machine learning models capable of accurately predicting device geometries and flow conditions required to generate stable aqueous-in-oil and oil-in-aqueous single and double emulsions from 15 to 250 m at rates up to 12000 Hz for different fluids commonly used in life sciences. Novel device geometries predicted by our models for as-yet-unseen fluids yield accurate predictions, establishing their generalizability. Finally, we generate an easy-to-use design automation tool that yield droplets within 3 m (< 8%) of the desired diameter, facilitating tailored droplet-based platforms for new applications and accelerating their utility in life sciences.

bioengineering↗