Detecting temporal changes in skin surrogate model using hyperspectral imaging

Recent research in biomedical imaging explores banana peels as a surrogate model for human skin due to shared biochemical traits, particularly the enzyme tyrosinase, linked to pigmentation and oxidative stress. This study explores the use of banana peels as a surrogate model for human skin due to their similar biochemical characteristics. We present a hyperspectral imaging (HSI) system analysing banana peels across the 440–900 nm range, detecting spectral changes linked to ripeness and oxidative stress. To reduce resource demands, we reconstructed 27-band hyperspectral data from RGB images. The results indicate significant potential for non-invasive diagnostic applications, particularly in dermatological research.

Graphic abstract

A preview of the hyperspectral (HS) image acquisition, capturing spectral information of selected regions of interest (ROIs) in skin model (banana peel) over multiple days. The dataset undergoes two key processes: (1) spectral reflectance analysis across different wavelengths over time, and (2) RGB-to-HSI mapping for detailed spectral representation.