RT Conference Proceedings
T1 Analyzing Digital Image by Deep Learning for Melanoma Diagnosis
A1 Thurnhofer Hemsi, Karl
A1 Domínguez-Merino, Enrique
K1 Congresos y conferencias
K1 Procesamiento de imágenes
K1 Melanoma
AB Image classi cation is an important task in many medicalapplications, in order to achieve an adequate diagnostic of di erent le-sions. Melanoma is a frequent kind of skin cancer, which most of themcan be detected by visual exploration. Heterogeneity and database sizeare the most important di culties to overcome in order to obtain a goodclassi cation performance. In this work, a deep learning based methodfor accurate classi cation of wound regions is proposed. Raw images arefed into a Convolutional Neural Network (CNN) producing a probabilityof being a melanoma or a non-melanoma. Alexnet and GoogLeNet wereused due to their well-known e ectiveness. Moreover, data augmentationwas used to increase the number of input images. Experiments show thatthe compared models can achieve high performance in terms of mean ac-curacy with very few data and without any preprocessing.
YR 2019
FD 2019-06-19
LK https://hdl.handle.net/10630/17841
UL https://hdl.handle.net/10630/17841
LA eng
NO Universidad de Málaga. Campus de Excelencia Internacional Andalucía Tech.
DS RIUMA. Repositorio Institucional de la Universidad de Málaga
RD 4 mar 2026