Spectral imaging of multi-color chromogenic dyes in pathological specimens

Area of Science:

• Pathology
• Biomedical Imaging
• Cytotechnology

Background:

• Cytopathology relies on accurate staining for diagnosis.
• Multi-color analysis of cytochemical dyes and enzyme precipitates is challenging.
• Digital imaging and spectroscopy offer potential for enhanced analysis.

Purpose of the Study:

• To investigate spectral imaging for multi-color analysis of cytochemical dyes and enzyme precipitates.
• To evaluate spectral imaging's reliability in classifying colors in cytopathological specimens.
• To explore spectral imaging's potential in improving pathological diagnosis and automated screening.

Main Methods:

• Utilized spectral imaging based on Fourier-transform spectroscopy and digital imaging.
• Developed pixel-by-pixel spectrum-based color classification algorithms.
• Applied methods to single-, double-, and triple-color in situ hybridization and immunocytochemical staining in cytopathological specimens.

Main Results:

• Spectral imaging unambiguously identified three chromogenic dyes in a single bright-field microscopic specimen.
• Demonstrated reliable pixel color recognition and classification for multi-color dyes.
• Showcased the ability to analyze serial microscopic fields using a spectral reference library.

Conclusions:

• Spectral imaging is a reliable and robust method for multi-color chromogenic dye analysis in cytopathology.
• Spectral imaging can increase simultaneously studied parameters in pathological diagnosis.
• This technique may improve quantitative data accuracy and solve segmentation issues in automated screening.