Hyperspectral Imaging for the Classification and Analysis of Drusen in Age-Related Macular Degeneration

Key Points

• The aim is to classify different types of drusen in age-related macular degeneration using hyperspectral imaging.
• Used hyperspectral imaging to capture spatial-spectral signatures of drusen types.
• Employed Spectral Angle Mapper (SAM) and random forest classifiers for analysis.
• Analyzed drusen mapping for noninvasive patient risk assessment.
• Identified distinct signatures for soft drusen, hard drusen, and reticular pseudodrusen (RPD).
• Automated detection methods achieved promising classification accuracy.
• Support for hyperspectral imaging as a tool for patient risk stratification.