Validation of a Microfluidic Device Prototype for Cancer Detection and Identification: Circulating Tumor Cells...

Area of Science:

• Biomedical Engineering
• Cancer Research
• Microfluidics

Background:

• Microfluidic devices (MDs) offer novel methods for detecting circulating tumor cells (CTCs).
• Current CTC detection methods often result in heterogeneous populations, requiring complex, costly, and time-consuming downstream processing for phenotype identification.
• There is a need for efficient and cost-effective approaches for CTC classification.

Purpose of the Study:

• To investigate the potential of a hyperuniform micropost microfluidic device (MD) for cost-effective and efficient circulating tumor cell (CTC) classification.
• To develop and validate a computational framework combining mathematical modeling and machine learning for CTC phenotype analysis.
• To assess the capability of the proposed approach in distinguishing between different CTC phenotypes based on cell trajectory.

Main Methods:

• Developed a cell-based modeling framework to simulate CTC dynamics within erythrocyte-laden plasma flow in a microfluidic channel.
• Utilized mathematical modeling of fluid-structure interactions to generate a comprehensive dataset of CTC trajectories.
• Employed machine learning techniques, specifically Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN), for CTC phenotype classification based on trajectory data.

Main Results:

• Generated a large dataset of simulated CTC trajectories representing two distinct CTC phenotypes.
• Demonstrated the effectiveness of CNN and RNN models in accurately classifying CTC phenotypes using trajectory data.
• Validated the potential of the hyperuniform micropost MD design for distinguishing between CTC phenotypes.

Conclusions:

• The hyperuniform micropost microfluidic device combined with machine learning analysis presents a promising, cost-effective strategy for CTC classification.
• This approach offers a potential avenue for improved early cancer detection through accurate identification of circulating tumor cell phenotypes.
• The study highlights the synergy between computational modeling and machine learning in advancing microfluidic-based diagnostics.