Machine learning for microfluidic design and control

McIntyre, David; Lashkaripour, Ali; Fordyce, Polly M.; Densmore, Douglas

doi:10.1039/d2lc00254j

Cited by 72 publications

(44 citation statements)

References 122 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In tissue engineering, droplet-based microfluidic systems are used to produce building blocks of artificial tissues and organs, such as shape-controlled micro particles and microfibers [ 19 ]. We also anticipate that new machine learning algorithms [ 63 , 64 , 65 ] have the potential to optimise the experimental parameters to improve single particle encapsulation efficiencies above the Poisson stochastic limit.…”

Section: Discussionmentioning

confidence: 99%

Viscoelastic Particle Encapsulation Using a Hyaluronic Acid Solution in a T-Junction Microfluidic Device

Jeyasountharan

Giudice

2023

Micromachines

View full text Add to dashboard Cite

The encapsulation of particles and cells in droplets is highly relevant in biomedical engineering as well as in material science. So far, however, the majority of the studies in this area have focused on the encapsulation of particles or cells suspended in Newtonian liquids. We here studied the particle encapsulation phenomenon in a T-junction microfluidic device, using a non-Newtonian viscoelastic hyaluronic acid solution in phosphate buffer saline as suspending liquid for the particles. We first studied the non-Newtonian droplet formation mechanism, finding that the data for the normalised droplet length scaled as the Newtonian ones. We then performed viscoelastic encapsulation experiments, where we exploited the fact that particles self-assembled in equally-spaced structures before approaching the encapsulation area, to then identify some experimental conditions for which the single encapsulation efficiency was larger than the stochastic limit predicted by the Poisson statistics.

show abstract

Section: Discussionmentioning

confidence: 99%

Viscoelastic Particle Encapsulation Using a Hyaluronic Acid Solution in a T-Junction Microfluidic Device

Jeyasountharan

Giudice

2023

Micromachines

View full text Add to dashboard Cite

show abstract

“…avoiding alignments). For instance, large language models (LLMs) 40,41 and other machine learning tools 42 show promise for streamlining the design of complex chip architectures and for generating many designs in parallel for high-throughput prototyping, while approaches like 2-photon lithography-based 3D printing 43 may offer increased flexibility, for instance for design of valves 44,45 and fluid routing in 3 dimensions. At present, the degree to which LOC modules can be composed into systems, and how much empirical testing will ultimately be required in design, remains an open question.…”

Section: Discussionmentioning

confidence: 99%

A workcell 1.0 for programmable and controlled operation of multiple fluidic chips in parallel

Ning

Bunke²,

Lietar³

et al. 2023

Preprint

View full text Add to dashboard Cite

We developed a versatile lab-on-chip (LOC) workcell that enables the design and automatic execution of experiments on LOC devices, improving how we establish, optimize, and productionalize LOC processes. Key features include direct docking and cooling of native laboratory tubes, programmable reagent mixing and dilutions, parallel operation of multiple chips, precise flowrate and pressure control, clogging detection and response, programmable microscope control, chip temperature regulation, and scheduled cleaning. All functionality is controlled seamlessly from an easy-to-write protocol file, and based on extensible hardware and software infrastructures to promote community development. To showcase the platform's use and versatility, we demonstrate a series of 5 different automated experiments at varying levels of complexity, executed across both Quake-valve and droplet microfluidic systems. In particular, the workcell was instructed to map the parameter regime that generates viable droplets, to allow a user to select diameters and production frequencies of interest for single bacterial cell encapsulation. Furthermore, three out of three days in a row, the platform successfully performed a complex 15.5h long experiment, integrating in a single automated protocol the full core workflow required by a typical protein-characterization lab: protein expression, purification, dilution generation, and quantitative binding characterization (generating 55296 images in the process). Experiments conducted through the workcell are easier to set up, offer increased control over experiment conditions and parameters, and can be heavily parallelized.

show abstract

“…channel clogging). These aspects have been extensively covered in a recent review by McIntyre et al 202 Lastly, ML can enhance the data analysis pipeline in preprocessing steps such as feature extraction or denoising, as well as in postprocessing steps including developing a prediction model and compressing multi-dimensional data for visualization.…”

Section: Label-free Cell Analysis Using Machine Learning Approachesmentioning

confidence: 99%

Label-free microfluidic cell sorting and detection for rapid blood analysis

et al. 2023

View full text Add to dashboard Cite

show abstract

Machine learning for microfluidic design and control

Abstract: In this review article, we surveyed the applications of machine learning in microfluidic design and microfluidic control.

Cited by 72 publications

References 122 publications

Viscoelastic Particle Encapsulation Using a Hyaluronic Acid Solution in a T-Junction Microfluidic Device

Viscoelastic Particle Encapsulation Using a Hyaluronic Acid Solution in a T-Junction Microfluidic Device

A workcell 1.0 for programmable and controlled operation of multiple fluidic chips in parallel

Label-free microfluidic cell sorting and detection for rapid blood analysis

Contact Info

Product

Resources

About