Real-time acquisition of quality verified nonstandardized color images for skin lesions risk assessment - A preliminary study

• Published in: 2014 18th International Conference on System Theory, Control and Computing (ICSTCC) pp. 199 - 204
• Main Authors: Udrea, Andreea, Lupu, Ciprian
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.10.2014
• Subjects: Indexes; Lesions; Malignant tumors; nonstandardized color images; real-time application; Real-time systems; Risk management; Shape; shape and texture analysis; Skin; skin lesions risk assessment
• DOI: 10.1109/ICSTCC.2014.6982415

Detection of melanoma in early stages is critical in terms of life expectancy of the patient and, also, in terms of treatment costs. In this context, we investigate the feasibility of implementing an algorithm that uses nonstandardized color images of skin lesions for risk assessment. In this paper we present a methodology for real-time acquisition of quality verified images using smartphones. We also test a series of shape and texture parameters to determine their relevance in the context of providing an accurate risk profile for an analyzed lesion. Our conclusion is that risk assessment with high accuracy is possible even when using nonstandardized color images taken with a mobile device (smartphone) camera.