top of page

Cancer Cell Tracker

CancerCellTracker_edited.jpg

This work was done by Dr. Qibing Jiang at University of Central Florida under the mentorship of Dr. Wei Zhang and me.

​

Time-lapse microscopy stands out as a transformative technique that captures live cell behavior through high-frequency imaging of ex vivo cultures. By documenting the dynamics of cells over time, this method offers valuable insights into how cancer cells react to various treatment strategies, thereby contributing to the advancement of precision oncology. Digital image processing algorithms typically depend on high-resolution images from limited, uniform cell line populations, which can restrict their applicability. Our proposed framework aims to enhance this process by enabling comprehensive tracking of patient-derived cancer cells, allowing us to quantify their behavior and viability in a high-throughput fashion, ultimately aiding in informed clinical decision-making.

​

​CancerCellTracker is a robust and user-friendly computational pipeline developed to analyze brightfield microscopy images of patient-derived tumor cells cultured ex vivo with bone marrow stromal cells, simulating the native tumor microenvironment. In this setup, cells are treated with 31 different drugs over six days, and CancerCellTracker performs automated cell detection in co-culture, tracks individual cells over time, and identifies cell death events by analyzing morphological and intensity changes in brightfield images—without relying on fluorescence labels. The pipeline was validated by comparing its cell death timing estimates to those obtained using ethidium homodimer, a fluorescent marker for dead cells, showing high concordance. Additionally, it was benchmarked against a state-of-the-art ImageJ-based algorithm, demonstrating comparable or improved performance, particularly in complex co-culture conditions. CancerCellTracker’s ability to accurately estimate the percentage of live cells in the presence of stromal cells, coupled with its scalability and non-invasive imaging requirements, makes it a powerful tool for high-throughput drug response profiling and dynamic, label-free phenotyping of primary cancer cells in clinically relevant settings.

bottom of page