Melanoma Prediction Using CNN Methods
Abstract
Melanoma is a type of skin cancer, and it is dangerous type of cancer, however, the melanoma can be found in the first stages of illness (Maia, L. et al., 2018). Moreover, as it is mentioned in (Diwan, T. et al., 2020), if the early stages of melanoma can be found as soon as possible, then it can increase the percentage of success treatment sufficiently. At the same time, with reference to (Esteva, A. et al., 2017), melanoma prediction at earliest stages is not easy task for dermatologist and Convolutional Neural Networks. As mentioned in the paper (Kousis, I. et al., 2022), “experienced specialist dermatologists had a success rate of only 60% until the invention of dermοscopic images”. Moreover, the images should have a good quality with corresponding to medical policy for medical images. Besides, the moles images are restricted in access and the photos of moles are not available in big amounts. That is why, trained model with maximized accuracy is a critical part of my program module for providing functionality for helping people to find any possible issues at first stages of developing symptoms.
Chapters:
- Part 1: Motivation
- Part 2: Melanoma Images Datasets
- Part 3: CNN Architectures
- Part 4: Machine learning frameworks
- Part 5: Jupiter notebook cloud environments
- Part 6: ML model serving
- Part 7: ML Model Serving Service Deployment Options
- Trained Model Preparation
- ML Model Serving
References
Maia, L.B., Lima, A., Pereira, R.M.P., Junior, G.B., de Almeida, J.D.S. and de Paiva, A.C., 2018, June. Evaluation of melanoma diagnosis using deep features. In 2018 25th international conference on systems, signals and image processing (IWSSIP) (pp. 1–4). IEEE.
[Accessed 25 May 2022]
Diwan, T., Shukla, R., Ghuse, E. and Tembhurne, J.V., 2022. Model hybridization & learning rate annealing for skin cancer detection. Multimedia Tools and Applications, pp.1–24.
[Accessed 28 May 2022]
Esteva, A., Kuprel, B., Novoa, R.A., Ko, J., Swetter, S.M., Blau, H.M. and Thrun, S., 2017. Dermatologist-level classification of skin cancer with deep neural networks. nature, 542(7639), pp.115–118.
[Accessed 30 May 2022]
Kousis, I., Perikos, I., Hatzilygeroudis, I. and Virvou, M., 2022. Deep Learning Methods for Accurate Skin Cancer Recognition and Mobile Application. Electronics, 11(9), p.1294.
[Accessed 29 May 2022]
