Um Modelo de Inteligência Artificial Para Detecção de Melanoma via Redes Neurais Convolucionais
Abstract
We propose an artificial intelligence (AI) model based on convolutional neural networks (CNN) for the task of melanoma and nevus classification. Due to its high risk of causing metastasis, melanoma is the most lethal type of skin cancer. However, its early detection significantly improves the probability of a favorable prognostic. Given this scenario, we intend to study the performance of the proposed CNN on the early detection of melanoma. By obtaining reasonable results in terms of specificity (0,96 ± 0.03) and sensibility (0,81 ± 0.02), we show that the proposed CNN model can be a valuable clinical diagnosis tool.
References
Brinker, T. J., Hekler, A., Enk, A. H., Klode, J., Hauschild, A., Berking, C., … e Schrüfer, P. (2019). Deep learning outperformed 136 of 157 dermatologists in a head-to-head dermoscopic melanoma image classification task. In European Journal of Cancer, 113, pages 47-54.
Haenssle, H. A., Fink, C., Schneiderbauer, R., Toberer, F., Buhl, T., Blum, A., ... e Zalaudek, I. (2018). Man against machine: diagnostic performance of a deep learning convolutional neural network for dermoscopic melanoma recognition in comparison to 58 dermatologists. In Annals of oncology, 29(8), pages 1836-1842.
Instituto Nacional do Câncer (2021). Tipos de Câncer: Melanoma. In Instituto Nacional do Câncer (Brasil). Disponível em: <https://www.inca.gov.br/tipos-de-cancer/cancerde-pele-melanoma>. Acesso em: 27 de maio de 2021.
Jordan, M. I., & Mitchell, T. M. (2015). Machine learning: Trends, perspectives, and prospects. In Science, 349, pages 255-260.
Murdoch, T. B. e Detsky, A. S. (2013). The inevitable application of big data to health care. In Jama, 309, pages 1351-1352.
Szegedy, C., Ioffe, S., Vanhoucke, V., & Alemi, A. (2017, February). Inception-v4, inception-resnet and the impact of residual connections on learning. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 31, No. 1).
Haenssle, H. A., Fink, C., Schneiderbauer, R., Toberer, F., Buhl, T., Blum, A., ... e Zalaudek, I. (2018). Man against machine: diagnostic performance of a deep learning convolutional neural network for dermoscopic melanoma recognition in comparison to 58 dermatologists. In Annals of oncology, 29(8), pages 1836-1842.
Instituto Nacional do Câncer (2021). Tipos de Câncer: Melanoma. In Instituto Nacional do Câncer (Brasil). Disponível em: <https://www.inca.gov.br/tipos-de-cancer/cancerde-pele-melanoma>. Acesso em: 27 de maio de 2021.
Jordan, M. I., & Mitchell, T. M. (2015). Machine learning: Trends, perspectives, and prospects. In Science, 349, pages 255-260.
Murdoch, T. B. e Detsky, A. S. (2013). The inevitable application of big data to health care. In Jama, 309, pages 1351-1352.
Szegedy, C., Ioffe, S., Vanhoucke, V., & Alemi, A. (2017, February). Inception-v4, inception-resnet and the impact of residual connections on learning. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 31, No. 1).
Published
2021-08-26
How to Cite
PESSANHA, Gabriel R. G.; CAMPOS, Eleanderson.
Um Modelo de Inteligência Artificial Para Detecção de Melanoma via Redes Neurais Convolucionais. In: REGIONAL SCHOOL OF APPLIED COMPUTING FOR HEALTH (ERCAS), 8. , 2021, São Paulo.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2021
.
p. 46-49.
DOI: https://doi.org/10.5753/ercas.2021.17436.
