Berkay M.Mergen E.H.Binici R.C.Bayhan Y.Gungor A.Okur E.Unay D.Türkan, Mehmet2023-06-162023-06-1620199.78E+12https://doi.org/10.1109/EBBT.2019.8741934https://hdl.handle.net/20.500.14365/35222019 Scientific Meeting on Electrical-Electronics and Biomedical Engineering and Computer Science, EBBT 2019 -- 24 April 2019 through 26 April 2019 -- 148870Melanoma which occurs with non-healing DNA degradation in melanocyte cells, is the most deadly type of skin cancers. Importantly, it can be identified for a treatment before it spreads to other tissues, i.e., early diagnosis. To identify, a specialist visually inspects whether the suspected lesion is melanoma or not. However, due to different education and experience levels of specialists or as a result of the patient not being in a facility that is specialized to this area, the problem of 'subjectivity' arises, and a good visual investigation accuracy may not always be achieved. Therefore, there is a significant need for automatic detection tools and systems. In this study, a method based on deep learning for automatic detection of melanoma from dermoscopic images is proposed. The developed system is tested with a large dataset and encouraging results are obtained. © 2019 IEEE.trinfo:eu-repo/semantics/closedAccessConvolutional neural networksDeep learningDermoscopyMelanomaSkin cancerBiomedical engineeringDeep learningDeep neural networksDiagnosisDiseasesElectronic medical equipmentLarge datasetNeural networksOncologyAutomatic DetectionConvolutional neural networkDermoscopic imagesDermoscopyMelanomaMelanoma detectionSkin cancersVisual investigationDermatologyConference Object10.1109/EBBT.2019.87419342-s2.0-85068548087