CNN optimization applied to Melanoma Diagnosis
Abstract
Melanoma is the most lethal skin cancer compared to others, however, patients have a high cure rate when diagnosed in their early stages. Exist several approaches to automatic diagnosis and detection proposed by different authors. However, the training of models in small and unbalanced databases presents several obstacles. Thus, this work in progress aims to apply the techniques of transfer of learning to training models capable of assisting in the diagnosis and screening of melanoma. Preliminary results showed that the use of synthetic data generation in conjunction with the fine-tuning of VGG16, showed crucial improvements, approximately 100% sensitivity and 93.75% specificity.
References
Image preprocessing. https://keras.io/preprocessing/image/. Accessed: 2018-12-05.
Baldi, P. and Sadowski, P. (2014). The dropout learning algorithm. Artificial intelligence, 210:78–122. http://dx.doi.org/10.1016/j.artint.2014.02.004
Chollet, F. et al. (2015). Keras. https://keras.io.
De Boer, P.-T., Kroese, D. P., Mannor, S., and Rubinstein, R. Y. (2005). A tutorial on the cross entropy method. Annals of operations research, 134(1):19–67. http://dx.doi.org/10.1007/s10479-005-5724-z
Klein, S., Pluim, J. P., Staring, M., and Viergever, M. A. (2009). Adaptive stochastic gradient descent optimisation for image registration. International journal of computer vision, 81(3):227. http://dx.doi.org/10.1007/s11263-008-0168-y
Mendonça, T., Ferreira, P. M., Marques, J. S., Marçal, A. R. S., and Rozeira, J. (2013). Ph2 - a dermoscopic image database for research and benchmarking. Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference, 2013:5437–40. http://dx.doi.org/10.1109/EMBC.2013.6610779
Mukkamala, M. C. and Hein, M. (2017). Variants of rmsprop and adagrad with logarithmic regret bounds. In Proceedings of the 34th International Conference on Machine Learning-Volume 70, pages 2545–2553. JMLR. org.
Simonyan, K. and Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556.
Soares, H. B. (2008). Análise e classificação de imagens de lesões da pele por atributos de cor, forma e textura utilizando máquina de vetor de suporte. PhD thesis, Universidade Federal do Rio Grande do Norte.
Xu, B., Wang, N., Chen, T., and Li, M. (2015). Empirical evaluation of rectified activations in convolutional network. arXiv preprint arXiv:1505.00853.
Zeiler, M. D. and Fergus, R. (2013). Stochastic pooling for regularization of deep convolutional neural networks. arXiv preprint arXiv:1301.3557.
Baldi, P. and Sadowski, P. (2014). The dropout learning algorithm. Artificial intelligence, 210:78–122. http://dx.doi.org/10.1016/j.artint.2014.02.004
Chollet, F. et al. (2015). Keras. https://keras.io.
De Boer, P.-T., Kroese, D. P., Mannor, S., and Rubinstein, R. Y. (2005). A tutorial on the cross entropy method. Annals of operations research, 134(1):19–67. http://dx.doi.org/10.1007/s10479-005-5724-z
Klein, S., Pluim, J. P., Staring, M., and Viergever, M. A. (2009). Adaptive stochastic gradient descent optimisation for image registration. International journal of computer vision, 81(3):227. http://dx.doi.org/10.1007/s11263-008-0168-y
Mendonça, T., Ferreira, P. M., Marques, J. S., Marçal, A. R. S., and Rozeira, J. (2013). Ph2 - a dermoscopic image database for research and benchmarking. Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference, 2013:5437–40. http://dx.doi.org/10.1109/EMBC.2013.6610779
Mukkamala, M. C. and Hein, M. (2017). Variants of rmsprop and adagrad with logarithmic regret bounds. In Proceedings of the 34th International Conference on Machine Learning-Volume 70, pages 2545–2553. JMLR. org.
Simonyan, K. and Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556.
Soares, H. B. (2008). Análise e classificação de imagens de lesões da pele por atributos de cor, forma e textura utilizando máquina de vetor de suporte. PhD thesis, Universidade Federal do Rio Grande do Norte.
Xu, B., Wang, N., Chen, T., and Li, M. (2015). Empirical evaluation of rectified activations in convolutional network. arXiv preprint arXiv:1505.00853.
Zeiler, M. D. and Fergus, R. (2013). Stochastic pooling for regularization of deep convolutional neural networks. arXiv preprint arXiv:1301.3557.
Published
2019-06-11
How to Cite
ROCHA, Carlos Vinicios Martins; MAIA, Lucas Bezerra ; JUNIOR, Geraldo Braz; DE ALMEIDA, João Dallyson Sousa; DE PAIVA, Anselmo Cardoso.
CNN optimization applied to Melanoma Diagnosis. In: BRAZILIAN SYMPOSIUM ON COMPUTING APPLIED TO HEALTH (SBCAS), 19. , 2019, Niterói.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2019
.
p. 336-341.
ISSN 2763-8952.
DOI: https://doi.org/10.5753/sbcas.2019.6272.
