Mitotic Index Determination on Live Cells From Label-Free Acquired Quantitative Phase Images Using a Supervised...

Area of Science:

• Biophysics
• Cell Biology
• Image Analysis

Background:

• Quantitative Phase Imaging (QPI) captures cellular dynamics non-invasively.
• Traditional cell cycle analysis often requires cell manipulation or labeling.

Purpose of the Study:

• To introduce and evaluate a new supervised autoencoder for classifying live cell populations.
• To assess the accuracy of QPI-based cell classification, particularly for mitotic cells.

Main Methods:

• Utilized Quantitative Phase Imaging (QPI) to acquire data from live cell populations.
• Applied a novel supervised autoencoder classification method, replacing the Douglas-Rachford method.
• Extracted quantitative features from QPI interferograms, including linear retardance and birefringence.

Main Results:

• Achieved high accuracy in classifying cells within the mitotic phase of the cell cycle.
• Demonstrated precise, non-invasive monitoring of cell growth without bias or cell labeling.
• Validated the efficacy of the autoencoder for live cell population analysis.

Conclusions:

• The supervised autoencoder with QPI features enables highly accurate, non-invasive live cell monitoring.
• This method facilitates precise tracking of cell growth and mitotic entry.
• Applicable to diverse studies involving cell growth analysis and treatment response evaluation.