Techniques for Malignant Melanoma Diagnosis: A Systematic Literature Review
Carlos I. Poclin Meza1, Kevin L. Monteza Corrales2, Lenis R. Wong Portillo3
1Carlos I. Poclin Meza, Software Engineering, Universidad Nacional Mayor de San Marcos, Lima, Peru.
2Kevin L. Monteza Corrales, Software Engineering, Universidad Nacional Mayor de San Marcos, Lima, Peru.
3Lenis R. Wong Portillo, Software Engineering, Universidad Nacional Mayor de San Marcos, Lima, Peru.
Manuscript received on August 01, 2020. | Revised Manuscript received on August 05, 2020. | Manuscript published on September 30, 2020. | PP: 80-86 | Volume-9 Issue-3, September 2020. | Retrieval Number: 100.1/ijrte.C4282099320 | DOI: 10.35940/ijrte.C4282.099320
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Malignant melanoma is the deadliest type of skin cancer. If melanoma detection and diagnosis is performed in its early stages, the probabilities of recovery and survival are higher. Dermoscopy is a manual method which is applied by doctors to diagnose this disease, but it strongly depends on the experience of the specialist who performs this skin assessment. Although, many proposals have been made for automated detection and diagnosis of malignant melanoma based on images processing, there are still improvement opportunities for melanoma diagnosis. This paper aims to identify the current status of the latest researches related to techniques for malignant melanoma diagnosis based on images analysis, considering the three research questions that have been elaborated for the systematic literature review: Q1) Which are the latest methods for malignant melanoma detection? Q2) Which systems for malignant melanoma diagnosis have been implemented in the last 5 years? And Q3) Which CAD systems for malignant melanoma detection have been developed? Furthermore, a cross-analysis of the outcome was performed. The results propose the implementation of systems using Inception V3 and the classifier Support Vector Machine, which achieved high accuracies in malignant melanoma diagnosis based on images processing.
Keywords: CAD Systems for Melanoma Diagnosis, CNN for Melanoma Detection, Dermoscopic Images Processing, Melanoma Detection, Support Vector Machine.