FalseColor-Python

FalseColor-Python converts two-channel fluorescence images into hematoxylin and eosin (H&E) color space to enable slide-free fluorescence-based digital pathology and quantitative comparison to traditional slide-based histology.

Key Features:

• Automated Intensity-Leveling: Automates intensity-leveling to compensate for intensity nonuniformities within and between specimens.
• Robust Color-Space Representation: Generates consistent H&E-like color-space mappings robust to variations in staining and imaging conditions across specimens.
• GPU-Accelerated Processing: Uses GPU-accelerated processing to handle large nondestructive 3D microscopy datasets efficiently.
• Versatility for Various Applications: Adapts the false-coloring methodology for other two-channel fluorescence-to-color-space rendering tasks.

Scientific Applications:

• Slide-free digital pathology: Produces H&E-like images from fluorescently imaged tissues to support slide-free digital pathology workflows.
• Nondestructive 3D microscopy adoption: Facilitates clinical adoption of nondestructive 3D microscopy by rendering diagnostically familiar H&E color space from fluorescence data.
• Cleared tissue OTLS imaging: Applied to cleared tissues imaged with open-top light-sheet (OTLS) microscopy to generate H&E-stained analogues for comparison with slide-based histology.
• Quantitative histology comparison: Enables quantitative comparison between fluorescence-derived H&E analogues and traditional slide-based histology.

Methodology:

The package applies an automated false-coloring algorithm that maps two fluorescence channels to H&E color space with intensity-leveling to correct nonuniformities and leverages GPU acceleration for large-dataset processing.