Project 2

Microfluidic Force Probes and Deep-Learning Segmentation for Intra-Tumoral Stress Quantification

(Dr. joseph chen, BE)

Cancer progression is highlighted by the emergence of aberrant mechanical features in the tumor microenvironment that present pathological signals to drive tumor malignancy. Alterations to extrinsic signals such as tumor ECM stiffness have been well described; however, the development of solid stresses within the tumor and its effect on tumor evolution is poorly understood. Recent reports have suggested that intra-tumoral stress guides cancer cell escape from tumors, highlighting an emergent mechanobiological driver of cancer progression; however, these aspects are difficult to investigate with standard in vitro tools and requires the development of advanced biophysical tools [1].

The goal of this project is to establish a microfluidic platform that generates deformable, monodisperse, fluorescent alginate microbeads to quantify compressive stresses generated within a growing tumorsphere (Figure 1). The REU student will design the platform, define optimized flow parameters, characterize bead properties, embed probes into 3D tumorspheres, quantify bead deformation over time, and extract compressive stress via finite element analysis. The student will also apply emerging AI based imaging analysis tools to accelerate bead parameter quantification [2]. These AI models include neural network segmentation and k-means clustering, foundational processes within machine learning pipelines, providing key AI perspectives and concepts for the REU student. Students interested in BE, ME, ChE, or Bio would be excellent candidates.

References

[1] Mohagheghian E, Luo J, Chen J, Chaudhary G, Chen J, Sun J, Ewoldt RH, Wang N. Quantifying compressive forces between living cell layers and within tissues using elastic round microgels. Nature communications. 2018 May 14;9(1):1878.

[2] Phan HA, Nguyen AT, Do Quang L, Thanh TB, Jen CP, Duc TC. Image-based machine learning quantitative evaluation of bead-cell binding interaction. Sensors and Actuators A: Physical. 2025 Feb 1;382:116123.