Description:
- Digital pathology
- Medical imaging
- Diagnostic software
- Life sciences tools
- Computational imaging
Abstract
USC researchers have developed a software-based imaging technology that improves visualization of hyperspectral autofluorescence data from unstained tissue samples. The technology converts complex spectral datasets into a compact, information-rich representation that enhances contrast and differentiation of tissue structures based on intrinsic autofluorescence signals. By enabling label-free, high-contrast imaging without chemical stains, this approach supports more objective and reproducible tissue analysis and is compatible with existing fluorescence imaging systems.
Benefit
- Enables label-free tissue visualization, eliminating chemical staining
- Enhances image contrast and interpretability of hyperspectral data
- Supports more objective and reproducible analysis
- Software-only solution compatible with existing imaging workflows
- Generates outputs suitable for automation and AI-based analysis
Market Application
This technology addresses a key limitation in hyperspectral autofluorescence imaging by enabling intuitive visualization of complex spectral data. It is applicable to clinical pathology and diagnostics, pharmaceutical and biotechnology research, life sciences imaging, and cosmetics and personal care product development. The solution can be deployed as standalone software or integrated into digital pathology and imaging platforms.
Publications
Transforming Hyperspectral Data into Insight: The DREAM Approach for Pathology,
Hong et al., published November 2025.
Other
Stage of Development
- Validated on hyperspectral autofluorescence datasets from unstained human tissue samples
- Performance benchmarked against conventional single-excitation imaging approaches
- Available for licensing