CTRL - a label-free artificial intelligence method for dynamic measurement of single-cell volume.

Area of Science:

• Cell Biology
• Biophysics
• Microscopy

Background:

• Accurate cell volume measurement is crucial for understanding cell growth control.
• Existing methods for mammalian cells are often labor-intensive, inflexible, and can damage cells.

Purpose of the Study:

• To introduce a novel, label-free technique for reconstructing cell topography and estimating cell volume.
• To develop a method that overcomes the limitations of current cell volume measurement techniques.

Main Methods:

• Utilizing a deep learning algorithm combined with the fluorescence exclusion method.
• Reconstructing cell topography and estimating cell volume from differential interference contrast (DIC) microscopy images.
• Applying the method to HT1080 fibrosarcoma cells for dynamic tracking.

Main Results:

• The Cell Topography Reconstruction Learner (CTRL) method achieves quantitative accuracy with minimal sample preparation.
• The technique is applicable to diverse biological and experimental conditions, enabling long-term single-cell volume tracking.
• HT1080 fibrosarcoma cells exhibit a positive correlation between cell size at birth and division (sizer).
• Cell size fluctuations in HT1080 cells show a reduction around 25% completion of the cell cycle.

Conclusions:

• CTRL offers a robust, efficient, and non-damaging approach for mammalian cell volume estimation.
• The method provides new insights into cell cycle dynamics and growth control mechanisms.
• This technique facilitates high-throughput and long-term monitoring of cell volume in various research settings.