Digital Holographic Microscopy for Phenotypic Profiling of Adherent Cells

Area of Science:

• Biophysics
• Cell Biology
• Computational Imaging

Background:

• Quantitative phase imaging, including digital holographic microscopy (DHM), enables non-invasive, high-resolution imaging of cellular morphology.
• Accurate reconstruction of microscale cell features like shape, height, and refractive index is crucial for understanding cell phenotypes.
• Phenotypic profiling using imaging data can reveal subtle cellular differences relevant to various biological and medical research areas.

Purpose of the Study:

• To detail a typical cell imaging experiment using digital holographic microscopy.
• To describe image analysis and machine learning workflows for phenotypic profiling.
• To demonstrate the classification and identification of typical and atypical cells within populations.

Main Methods:

• Digital holographic microscopy (DHM) for quantitative phase imaging of cells.
• Image processing techniques for extracting microscale features (shape, height, refractive index).
• Machine learning algorithms for single and multi-class classification of cell phenotypes.

Main Results:

• DHM successfully reconstructed detailed microscale cellular features.
• Image analysis pipeline effectively extracted relevant phenotypic parameters.
• Machine learning models achieved accurate classification of typical and atypical cells.

Conclusions:

• Digital holographic microscopy is a powerful tool for quantitative cell analysis and phenotypic profiling.
• Integrated image analysis and machine learning enable robust identification of cellular phenotypes.
• This approach facilitates the distinction between typical and atypical cells, aiding biological research.