Performance Analysis of the Lesion Area Clustering Process From Dermatological Images in Embedded Systems

Embedded systems that come into our lives with the developing technology have started to be preferred in the field of health, as in many other fields. In this paper, we investigated the performance analysis of the embedded systems devices usage for medical purposes. Two different development boards which are BeagleBone Black Wireless and Raspberry Pi 3 were used for this purpose. The K-means algorithm were used on the development boards for clustering the lesion areas on dermatological images from Melanoma-Naevus Dataset. It was concluded that the best result was obtained by BeagleBone Black Wireless development board, according to the performance analysis.