Abstract
Non-melanoma skin tumors, mainly basal cell carcinoma and squamous cell carcinoma, are the most common human cancers. Early detection and discrimination of skin tumors is of paramount importance to decision making and treatment. The main treatment for these skin tumors is surgical excision, but its extent is strongly influenced by the preoperative diagnosis. This study presents a new method for skin tumor discrimination based on tumor oxygenation levels extracted from hyperspectral images. Hyperspectral images of 16 skin tumors (four actinic keratoses, six basal cell carcinomas, six squamous cell carcinomas) were obtained prior excision and pathological diagnosis. The concentrations of oxyhemoglobin, deoxyhemoglobin and oxygen saturation levels were measured from hyperspectral images using an algorithm based on the modified Beer–Lambert law. The results were compared with pathology diagnosis. The results revealed that there were statistically significant differences in the mean oxyhemoglobin concentrations and oxygen saturation levels between actinic keratoses and basal cell carcinomas, between basal cell carcinomas and squamous cell carcinomas and between actinic keratoses and squamous cell carcinomas. Deoxyhemoglobin concentrations were not statistically different between the two carcinoma types but were different between carcinomas and actinic keratoses. In conclusion, the proposed method proved that it could be used as a reliable non-invasive diagnostic tool for differentiating benign from malignant skin tumors with the possibility of extending its applications to other medical research areas.
This is a visual representation of the abstract.
Keywords
Get full access to this article
View all access options for this article.