CASTLE

CASTLE automates quantitative analysis of cell adhesion and morphology from phase-contrast microscopy (PCM) images using image processing, statistical analysis, and machine learning.

Key Features:

• Automated Cell Counting and Morphology Analysis: Automated cell counting and morphology assessment in PCM images with validation against manual counts and segmentation.
• PCM-specific Image Processing: Workflow and image processing methods tailored to the challenges of phase-contrast microscopy images.
• Statistical and Machine Learning Techniques: Uses statistical analysis and machine learning methods including principal component analysis (PCA) and k-means clustering to extract complex morphological patterns.

Scientific Applications:

• Adherent cell analysis: Validation and analysis of adherent cell types including monocytes, neutrophils, and platelets.
• Investigation of molecular effects on adhesion: Quantitative assessment of how molecules affect cell adhesion, exemplified by studies showing Galectin-9 influences leukocyte adhesion.
• Non-intrusive monitoring of cellular dynamics: Analysis of cell adhesion dynamics and mechanics using non-destructive PCM imaging.

Methodology:

Computational workflow for PCM images comprising image processing and segmentation, automated cell counting, statistical analyses, and machine learning methods such as PCA and k-means clustering.