Machine learning enables design automation of microfluidic flow-focusing droplet generation.

Machine learning enables design automation of microfluidic flow-focusing droplet generation.
复制标题

DOI:
10.1038/s41467-020-20284-z
复制
发表时间:
2021-01-04
影响因子:
16.6
通讯作者:
Densmore D
Densmore D
中科院分区:
综合性期刊1区
文献类型:
--
作者:
Lashkaripour A;Rodriguez C;Mehdipour N;Mardian R;McIntyre D;Ortiz L;Campbell J;Densmore D

文献摘要

参考文献

被引文献

相似文献

基于液滴的微流体设备具有巨大的潜力,可以成为生命科学应用中现有筛选平台的廉价替代品,例如酶发现和早期癌症检测。然而,缺乏对液滴生成的预测性理解,使得基于液滴的平台的设计成为一个迭代和资源密集型的过程。我们提出了一种基于网络的工具 DAFD,它可以预测流量聚焦液滴发生器的性能并实现设计自动化。我们利用机器学习算法来预测液滴直径和速率,平均绝对误差小于 10 μm 和 20 Hz。该工具可提供用户指定的性能,误差范围在所需直径和速率的 4.2% 至 11.5% 范围内。我们证明了社区可以扩展 DAFD 以支持其他流体组合,而不需要广泛的机器学习知识或大规模数据集。该工具将减少对微流体专业知识和设计迭代的需求,并促进微流体在生命科学中的采用。液滴生成装置是许多微流体应用的核心,但很难针对特定情况进行定制。拉什卡里普尔等人。展示如何通过将快速原型制作与数据驱动的机器学习策略相结合来大大简化设计定制。
Droplet-based microfluidic devices hold immense potential in becoming inexpensive alternatives to existing screening platforms across life science applications, such as enzyme discovery and early cancer detection. However, the lack of a predictive understanding of droplet generation makes engineering a droplet-based platform an iterative and resource-intensive process. We present a web-based tool, DAFD, that predicts the performance and enables design automation of flow-focusing droplet generators. We capitalize on machine learning algorithms to predict the droplet diameter and rate with a mean absolute error of less than 10 μm and 20 Hz. This tool delivers a user-specified performance within 4.2% and 11.5% of the desired diameter and rate. We demonstrate that DAFD can be extended by the community to support additional fluid combinations, without requiring extensive machine learning knowledge or large-scale data-sets. This tool will reduce the need for microfluidic expertise and design iterations and facilitate adoption of microfluidics in life sciences. Devices for droplet generation are at the heart of many microfluidic applications but difficult to tailor for specific cases. Lashkaripour et al. show how design customization can greatly be simplified by combining rapid prototyping with data-driven machine learning strategies.
DOI: 10.1039/c8lc01394b
发表时间: 2019-05-21
期刊: LAB ON A CHIP
影响因子: 6.1
作者:
Chu, Albert;Du Nguyen;Giera, Brian
通讯作者: Giera, Brian
DOI: 10.1038/nbt.3880
发表时间: 2017-07
影响因子: 46.9
作者:
Lan F;Demaree B;Ahmed N;Abate AR
通讯作者: Abate AR
DOI: 10.1016/j.ces.2004.04.006
发表时间: 2004-08-01
影响因子: 4.7
作者:
Cramer, C;Fischer, P;Windhab, EJ
通讯作者: Windhab, EJ
DOI: 10.1016/j.jbiomech.2015.11.031
发表时间: 2016-07-26
影响因子: 2.4
作者:
Faustino, Vera;Catarino, Susana O.;Minas, Graca
通讯作者: Minas, Graca
DOI: 10.1007/s10404-018-2048-2
发表时间: 2018-03-01
影响因子: 2.8
作者:
Lashkaripour, Ali;Silva, Ryan;Densmore, Douglas
通讯作者: Densmore, Douglas