Skip to main content
CISTI'2020 - 15th Iberian Conference on Information Systems and Technologies

Full Program »

Extended Model for the Early Skin Cancer Detection Using Image Processing

In this research paper, we proposed an extended model for the early detection of skin cancer. This model has automatized the process of dermatological care by reducing waiting times when obtaining preliminary results. The function of the dermatoscope has been digitized by considering the ABCD rule in dermatology through the use of Smartphones and the use of magnifying lenses in these mobile devices. This model is about an interpretation from the cell phone lens, which captures an image and stores it in the cloud, to be then processed. It provides a probability of suffer from skin cancer, which is confirmed by the doctor's diagnosis. This model is made up of four elements: 1) the patient is followed up by a general practitioner and a nurse previously trained, the health centre must have Wifi connection or mobile network to record data and the image of the skin lesion that will be analysed. 2) The image will be in the cloud storage, which at the same time feeds an exclusive access website of dermatologists. 3) Images are analysed in real time using an image recognition service provided by IBM that can be found in the cloud. 4) The result of the image processing is visualized by the doctor who makes a remote diagnosis. This extended technological model was checked in a group of 15 patients, who were diagnosed with skin cancer in early stage, in a network of clinics in Lima, Peru. The results were successful in a 97.5% of all the skin lesions analysed, because fifteen of them showed a percentage of probability between 90% and 95%.

Jimmy Armas
Universidad Peruana de Ciencias Aplicadas
Peru

Jonathan Campos
Universidad Peruana de Ciencias Aplicadas
Peru

Emily De la Cruz
Universidad Peruana de Ciencias Aplicadas
Peru

Leonor Gutiérrez
Universidad Autónoma de Coahuila
Mexico

 


Powered by OpenConf®
Copyright ©2002-2018 Zakon Group LLC