Advancing dermatological diagnosis: Development of a hyperspectral dermatoscope for enhanced skin imaging

TL;DR

Preliminary results from 15 individuals and 160 recorded skin images demonstrate the potential of the Hyperscope in identifying and characterizing various skin conditions, offering a promising avenue for non-invasive skin evaluation and a platform for future research in dermatology-related hyperspectral imaging.

Abstract

Clinical dermatology necessitates precision and innovation for efficient diagnosis and treatment of various skin conditions. This paper introduces the development of a cutting-edge hyperspectral dermatoscope (the Hyperscope) tailored for human skin analysis. We detail the requirements to such a device and the design considerations, from optical configurations to sensor selection, necessary to capture a wide spectral range with high fidelity. Preliminary results from 15 individuals and 160 recorded skin images demonstrate the potential of the Hyperscope in identifying and characterizing various skin conditions, offering a promising avenue for non-invasive skin evaluation and a platform for future research in dermatology-related hyperspectral imaging.

Figures (11)

• Figure 1: Measurement of the counts using a halogen illumination with the sensor and the filter used. The brightness used for the respective channel is not identical, but it is clearly recognisable that the FWHM increases for longer wavelengths due to dispersion.
• Figure 2: Measured spectrum of the broadband LED lighting used consisting of several white LEDs and two different types of infrared LEDs. The two peaks of the infrared LEDs are at 910 nm and 970 nm.
• Figure 3: a) Close-up of the developed hyperspectral dermatoscope. b) Image of the entire setup. On the left is the control and power supply for the lighting and camera. The docking station with calibration target can be seen at the top right.
• Figure 4: A pigmented lesion located on the torso viewed at different wavelengths. The images show the lesion with steps of 50 nm. Except for 3 and 4, which show the wavelengths 540 nm and 570 nm respectively, which represent the hemoglobin absorption peaks. With increasing wavelength in the visible range, the increased reflectance can be clearly recognised in the image.
• Figure 5: Median reflectance of the 'Skin' and 'Lesion' class of all body parts and all patient ids. For each measurement, the mean value was determined over an area of $3\times3$ pixels around the annotated location. The median is shown here since it preserves spectral features better than the mean.