Algorithm for the Analysis of Pigment Network Characteristics in Diagnosing Melanoma

Abstract: An algorithm for analyzing the characteristics of the pigment network of skin neoplasms is proposed. It is based on the assessment of the deviation coefficient of the average lengths of the pigment network segments in the local areas of the neoplasm from the average value of the lengths of the pigment network segments throughout the area of the neoplasm. The use of the algorithm makes it possible to distinguish the typical pigment network from an atypical one. An atypical pigment network is a core feature in identifying early melanoma. The algorithm can be used in automated systems to support medical decision-making in the diagnosis of skin neoplasms. © 2021, Pleiades Publishing, Ltd.