Abstract
Melanoma is one of the most treacherous forms of cancer, and its early detection is paramount for the survival rate. It is caused by anomalous multiplication of skin cells, giving that area an unusual color. In this paper, we present a method for melanoma classification based on Efficient Nets, squeeze and excitation models, attention mechanisms, and ensembling. In this work, we consider different image sizes are utilized for different Efficient Nets, to act as the backbone of our models and this plays an important role in our proposed method. The feature maps are then passed to convolution layers with a Squeeze and Excitation structure, further followed by an attention mechanism. A separate branch for patient-level data is also used to improve the results. They are combined using two novel ensemble techniques: the majority mean ensemble and the absolute correlation ensemble, to give a final prediction. We also compare our results with the basic mean ensemble to prove their superiority.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
J.F. Thompson, R.A. Scolyer, R.F. Kefford, Cutaneous melanoma. The Lancet 365(9460), 687–701 (2005). https://doi.org/10.1016/S0140-6736(05)17951-3. http://www.sciencedirect.com/science/article/pii/S0140673605179513
A.J. Miller, M.C. Mihm, Melanoma. New England J. Med. 355(1), 51–65 (2006). https://doi.org/10.1056/NEJMra052166
M. Tan, Q.V. Le, Efficientnet: rethinking model scaling for convolutional neural networks. CoRR (2019). http://arxiv.org/abs/1905.11946
H. Choi, K. Cho, Y. Bengio, Fine-grained attention mechanism for neural machine translation. Neurocomputing 284, 171–176 (2018). https://doi.org/10.1016/j.neucom.2018.01.007. http://www.sciencedirect.com/science/article/pii/S0925231218300225
J. Hu, L. Shen, G. Sun, Squeeze-and-excitation networks, in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2018)
P. Ramachandran, B. Zoph, Q.V. Le, Searching for activation functions (2017)
T.T. Wong, Performance evaluation of classification algorithms by k-fold and leave-one-out cross validation. Pattern Recogn. 48(9), 2839–2846 (2015). https://doi.org/10.1016/j.patcog.2015.03.009. http://www.sciencedirect.com/science/article/pii/S0031320315000989
J. Hron, A.G. de Matthews, Z. Ghahramani, Variational gaussian dropout is not bayesian (2017)
Q. Xie, M.T. Luong, E. Hovy, Q.V. Le, Self-training with noisy student improves imagenet classification (2020)
S. Gulati, R.K. Bhogal, Detection of Malignant Melanoma Using Deep Learning, in Advances in Computing and Data Sciences (Singapore, Springer Singapore, 2019), pp. 312–325
H. Nahata, S.P. Singh, Deep Learning Solutions for Skin Cancer Detection and Diagnosis (Springer International Publishing, Cham, 2020), pp. 159–182. https://doi.org/10.1007/978-3-030-40850-3_8
J. Amin, A. Sharif, N. Gul, M.A. Anjum, M.W. Nisar, F. Azam, S.A.C. Bukhari, Integrated design of deep features fusion for localization and classification of skin cancer. Pattern Recogn. Lett. 131, 63–70 (2020). https://doi.org/10.1016/j.patrec.2019.11.042. http://www.sciencedirect.com/science/article/pii/S0167865519303630
Gessert, N., Nielsen, M., Shaikh, M., Werner, R., Schlaefer, A.: Skin lesion classification using ensembles of multi-resolution efficient nets with meta data (2019)
T. DeVries, G.W. Taylor, Improved Regularization of Convolutional Neural Networks with Cutout (2017)
R. Takahashi, T. Matsubara, K. Uehara, Ricap: random image cropping and patching data augmentation for deep cnns. (PMLR, 2018), pp. 786–798. http://proceedings.mlr.press/v95/takahashi18a.html
S. Yun, D. Han, S.J. Oh, S. Chun, J. Choe, Y. Yoo, Cutmix: regularization strategy to train strong classifiers with localizable features (2019)
T.Y. Hsiao, Y.C. Chang, H.H. Chou, C.T. Chiu, Filter-based deep-compression with global average pooling for convolutional networks. J. Syst. Architecture 95, 9–18 (2019). https://doi.org/10.1016/j.sysarc.2019.02.008. http://www.sciencedirect.com/science/article/pii/S1383762118302340
J. Fan, S. Upadhye, A. Worster, Understanding receiver operating characteristic (roc) curves. Canad. J. Emerg. Med. 8(1), 19–20 (2006). https://doi.org/10.1017/S1481803500013336
C. Shorten, T. Khoshgoftaar, A survey on image data augmentation for deep learning. J. Big Data 6, 1–48 (2019)
A.C. Marreiros, J. Daunizeau, S.J. Kiebel, K.J. Friston, Population dynamics: variance and the sigmoid activation function. NeuroImage 42(1), 147–157 (2008). https://doi.org/10.1016/j.neuroimage.2008.04.239. http://www.sciencedirect.com/science/article/pii/S1053811908005132
Y. Cui, M. Jia, T.Y. Lin, Y. Song, S. Belongie, Class-balanced loss based on effective number of samples, in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (2019)
T.Y. Lin, P. Goyal, R. Girshick, K. He, P. Dollár, Focal loss for dense object detection (2018)
D.P. Kingma, J. Ba, Adam: a method for stochastic optimization (2017)
V. Rotemberg, N. Kurtansky, B. Betz-Stablein, L. Caffery, E. Chousakos, N. Codella, M. Combalia, S. Dusza, P. Guitera, D. Gutman, A. Halpern, Kittler, H., K. Kose, S. Langer, K. Lioprys, J. Malvehy, S. Musthaq, J. Nanda, O. Reiter, G. Shih, A. Stratigos, P. Tschandl, J. Weber, H.P. Soyer, A patient-centric dataset of images and metadata for identifying melanomas using clinical context (2020). https://doi.org/10.34970/2020-ds01
N. Codella, V. Rotemberg, P. Tschandl, M.E. Celebi, S. Dusza, D. Gutman, B. Helba, A. Kalloo, K. Liopyris, M. Marchetti, H. Kittler, A. Halpern, Skin Lesion Analysis Toward Melanoma Detection 2018: A Challenge Hosted by the International Skin Imaging Collaboration (ISIC) (2019)
N.C.F. Codella, D. Gutman, M.E. Celebi, B. Helba, M.A. Marchetti, S.W. Dusza, A. Kalloo, K. Liopyris, N. Mishra, H. Kittler, A. Halpern, Skin Lesion Analysis Toward Melanoma Detection: A Challenge at the 2017 International Symposium on Biomedical Imaging (isbi), Hosted by the International Skin Imaging Collaboration (ISIC) (2018)
P. Tschandl, The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions (2018). https://doi.org/10.7910/DVN/DBW86T
Caruana, R., Munson, A., Niculescu-Mizil, A.: Getting the most out of ensemble selection, in Sixth International Conference on Data Mining (ICDM’06), pp. 828–833 (2006)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Agarwal, V., Jhalani, H., Singh, P., Dixit, R. (2022). Classification of Melanoma Using Efficient Nets with Multiple Ensembles and Metadata. In: Tiwari, R., Mishra, A., Yadav, N., Pavone, M. (eds) Proceedings of International Conference on Computational Intelligence. Algorithms for Intelligent Systems. Springer, Singapore. https://doi.org/10.1007/978-981-16-3802-2_8
Download citation
DOI: https://doi.org/10.1007/978-981-16-3802-2_8
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-16-3801-5
Online ISBN: 978-981-16-3802-2
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)