Zhang Naiyin, Liang Kaicong, Liu Zhenya, Sun Taotao, Wang Junchao
School of Automation, Hangzhou Dianzi University, Hangzhou 310018, China.
Key Laboratory of RF Circuits and Systems, Ministry of Education, Hangzhou Dianzi University, Hangzhou 310018, China.
Micromachines (Basel). 2022 Nov 28;13(12):2100. doi: 10.3390/mi13122100.
Microfluidics has shown great potential in cell analysis, where the flowing path in the microfluidic device is important for the final study results. However, the design process is time-consuming and labor-intensive. Therefore, we proposed an ANN method with three dense layers to analyze particle trajectories at the critical intersections and then put them together with the particle trajectories in straight channels. The results showed that the ANN prediction results are highly consistent with COMSOL simulation results, indicating the applicability of the proposed ANN method. In addition, this method not only shortened the simulation time but also lowered the computational expense, providing a useful tool for researchers who want to receive instant simulation results of particle trajectories.
微流控技术在细胞分析中已展现出巨大潜力,其中微流控装置中的流动路径对最终研究结果至关重要。然而,设计过程既耗时又费力。因此,我们提出了一种具有三个全连接层的人工神经网络(ANN)方法,用于分析关键交叉点处的粒子轨迹,然后将其与直通道中的粒子轨迹整合在一起。结果表明,ANN预测结果与COMSOL模拟结果高度一致,这表明所提出的ANN方法具有适用性。此外,该方法不仅缩短了模拟时间,还降低了计算成本,为想要获得粒子轨迹即时模拟结果的研究人员提供了一个有用的工具。