Background: Cutaneous melanoma is a melanocytic skin tumour, which has very poor prognosis while it is highly resistant to treatment and tends to metastasize. Thickness of melanoma is one of the most important biomarker for stage of disease, prognosis and surgery planning. In this study, we hypothesized that the analysis of spectrophotometric (SIAscope) images can provide the information about skin tumour thickness.
Methods: The intensity of blood displacement, "erythematous blush", collagen holes, intensity of collagen, dermal and epidermal melanin were estimated in SIAgraphs. Tumour thicknesses were evaluated non-invasively in ultrasound images before excision. The diagnosis and Breslow index of each tumour were evaluated during routine histological examination.
Results: The logistic regression analysis of two thicknesses groups of melanocytic tumours (≤1 mm, n = 72 and >1 mm, n = 30), using six SIAscopic features lead to achieve the areas under the ROC curves of 0.9 and 0.96 respectively. Overall the sensitivity and specificity of SIAscopy observed in this study is 81.4% and 86.4% respectively.
Conclusion: The features of SIAgraphs individually are not enough specific for melanoma diagnosis with different thickness. Promising results were observed for differentiation of melanocytic skin tumour, using all 6 SIAscopic features, which correspond to the distribution, location and concentration of skin chromophores.
Keywords: high-frequency ultrasound; melanocytic skin tumour; spectrophotometric intracutaneous analysis; tumour thickness.
© 2018 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd.