Enhanced Robustness in Automated Skin Cancer Detection via a CNN-Based Ensemble Classifier

Abstract

Malignant skin lesions, commonly referred to as skin cancer, represent a leading cause of global mortality. The integration of advanced deep learning methodologies facilitates the early diagnosis of these lesions, thereby offering a vital pathway to decrease associated mortality rates. For the automated categorization of dermoscopic images, deep convolutional neural networks (CNNs) have consistently exhibited performance superior to that of traditional machine learning algorithms. The current research focuses on the development of a highly precise ensemble classifier built upon foundational CNN architectures for the purpose of classifying skin lesions. To ensure the broad generalizability of the proposed model, strategic enhancements in diversity were implemented to bolster both systemic resilience and accuracy. Specifically, this necessary variance was introduced at both the dataset and the algorithmic classifier levels. The comprehensive ISIC 2018 challenge dataset, encompassing more than 13,000 dermoscopic images, was utilized across the training, validation, and testing phases. Experimental outcomes indicate that the developed ensemble framework attained an accuracy rate of 90.15%, successfully surpassing baseline CNN configurations and comparable contemporary classifiers in terms of overall accuracy and robust performance.

Keywords: Deep Learning Convolutional Neural Networks Ensemble Classification Dermoscopy Skin Lesion Diagnosis

References

S. K. Datta, M. A. Shaikh, S. N. Srihari, and M. Gao, “Soft attention improves skin cancer classification performance,” in Interpretability of Machine Intelligence in Medical Image Computing, and Topological Data Analysis and Its Applications for Medical Data, M. Reyes, P. Henriques Abreu, J. Cardoso, M. Hajij, G. Zamzmi, P. Rahul, and L. Thakur, Eds., Cham, Switzerland : Springer, 1007, pp. 13–23.

A. N. Omeroglu, H. M. A. Mohammed, E. A. Oral, and S. Aydin, “A novel soft attention-based multi-modal deep learning framework for multi-label skin lesion classification,” Eng. Appl. Artif. Intell., vol. 120, Apr. 2023, Art. no. 105897.

P. Choudhary, J. Singhai, and J. S. Yadav, “Skin lesion detection based on deep neural networks,” Chemometric Intell. Lab. Syst., vol. 230, Nov. 2022, Art. no. 104659.

G. Cirrincione, S. Cannata, G. Cicceri, F. Prinzi, T. Currieri, M. Lovino, C. Militello, E. Pasero, and S. Vitabile, “Transformer-based approach to melanoma detection,” Sensors, vol. 23, no. 12, p. 5677, Jun. 2023.

L. I. Kuncheva, Combining Pattern Classifiers: Methods and Algorithms. Hoboken, NJ, USA : Wiley, 2014.

N. Codella, V. Rotemberg, P. Tschandl, M. Emre Celebi, S. Dusza, D. Gutman, B. Helba, A. Kalloo, K. Liopyris, M. Marchetti, H. Kittler, and A. Halpern, “Skin lesion analysis toward melanoma detection 2018: A challenge hosted by the international skin imaging collaboration (ISIC),” 2019, arXiv:1902.03368.

ISIC Challenge2018, ISIC, 2018.

F. Zhuang, Z. Qi, K. Duan, D. Xi, Y. Zhu, H. Zhu, H. Xiong, and Q. He, “A comprehensive survey on transfer learning,” Proc. IEEE, vol. 109, no. 1, pp. 43–76, Jan. 2021.

Z. Liu, H. Mao, C.-Y. Wu, C. Feichtenhofer, T. Darrell, and S. Xie, “A ConvNet for the 2020s,” in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR), Jun. 2022, pp. 11966–11976.

J. Hu, L. Shen, and G. Sun, “Squeeze-and-excitation networks,” in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit., Jun. 2018, pp. 7132–7141.

S. Mekruksavanich and A. Jitpattanakul, “Deep residual network for smartwatch-based user identification through complex hand movements,” Sensors, vol. 22, no. 8, p. 3094, Apr. 2022.

M. Tan and Q. Le, “EfficientNet: Rethinking model scaling for convolutional neural networks,” in Proc. Int. Conf. Mach. Learn., 2019, pp. 6105–6114.

G. Huang, Z. Liu, L. Van Der Maaten, and K. Q. Weinberger, “Densely connected convolutional networks,” in Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR), Jul. 2017, pp. 4700–4708.

PyTorch. ( 2024 ). DenseNet—PyTorch Hub. Accessed: Jul. 1, 2024. [Online]. Available: https://pytorch.org/hub/pytorch_vision_densenet/

X. Glorot and Y. Bengio, “Understanding the difficulty of training deep feedforward neural networks,” in Proc. 13th Int. Conf. Artif. Intell. Statist., 2010, pp. 249–256.

ISIC. ( 2018 ). ISIC 2018 Challenge: Leaderboards. Accessed: Jan. 2024.

K. Nakai, Y.-W. Chen, and X.-H. Han, “Enhanced deep bottleneck transformer model for skin lesion classification,” Biomed. Signal Process. Control, vol. 78, Sep. 2022, Art. no. 103997.

C. Mingang and W. Chen. ( 2018 ). Skin Disease Classification By Deep Network. [Online]. Available: https://isic-challenge-stade.s3.amazonaws.com/aa13d171-30f5-4622-b095-1a381f0d19d5/Skin_Disease_Classification_by_Deep_Network.pdf

Y. Xue, L. Gong, W. Peng, X. Huang, and Y. Zheng. ( 2018 ). Automatic Skin Lesion Analysis With Deep Networks. [Online]. Available: https://isic-challenge-stade.s3.amazonaws.com/4c9b36c3-161b-4b50-a820-92b85462f844/ISIC2018.pdf

A. H. Shahin, A. Kamal, and M. A. ElAttar. ( 2018 ). Automatic Skin Lesions Diagnosis With Deep Neural Networks. [Online]. Available: https://isic-challenge-stade.s3.amazonaws.com/1b91ec7e-1ab4-4440-ace0-4b30f219f8c7/Task3_article.pdf

T. Majtner, B. Bajić, S. Yildirim, J. Y. Hardeberg, J. Lindblad, and N. Sladoje, “Convolutional neural network based classification of dermoscopic images by transfer learning: ISIC 2018 challenge,” ISIC Challenge Website Leaderboard Standing, Tech. Rep., 2022.

S. Maqsood and R. Damaševičius, “Multiclass skin lesion localization and classification using deep learning based features fusion and selection framework for smart healthcare,” Neural Netw., vol. 160, pp. 238–258, Mar. 2023.

P. N. Srinivasu, J. G. Sivasai, M. F. Ijaz, A. K. Bhoi, W. Kim, and J. J. Kang, “Classification of skin disease using deep learning neural networks with MobileNet V2 and LSTM,” Sensors, vol. 21, no. 8, p. 2852, Apr. 2021.

B. Shetty, R. Fernandes, A. P. Rodrigues, R. Chengoden, S. Bhattacharya, and K. Lakshmanna, “Skin lesion classification of dermoscopic images using machine learning and convolutional neural network,” Sci. Rep., vol. 12, no. 1, p. 18134, Oct. 2022.

M. Wei, Q. Wu, H. Ji, J. Wang, T. Lyu, J. Liu, and L. Zhao, “A skin disease classification model based on DenseNet and ConvNeXt fusion,” Electronics, vol. 12, no. 2, p. 438, Jan. 2023.

I. Kousis, I. Perikos, I. Hatzilygeroudis, and M. Virvou, “Deep learning methods for accurate skin cancer recognition and mobile application,” Electronics, vol. 11, no. 9, p. 1294, Apr. 2022.

M. A. Elashiri, A. Rajesh, S. N. Pandey, S. K. Shukla, S. Urooj, and A. Lay-Ekuakille, “Ensemble of weighted deep concatenated features for the skin disease classification model using modified long short term memory,” Biomed. Signal Process. Control, vol. 76, Jul. 2022, Art. no. 103729.

A. Dosovitskiy, L. Beyer, A. Kolesnikov, D. Weissenborn, X. Zhai, T. Unterthiner, M. Dehghani, M. Minderer, G. Heigold, S. Gelly, J. Uszkoreit, and N. Houlsby, “An image is worth 16 × 16 words: Transformers for image recognition at scale,” 2020, arXiv:2010.11929.

Z. Qin, Z. Liu, P. Zhu, and Y. Xue, “A GAN-based image synthesis method for skin lesion classification,” Comput. Methods Programs Biomed., vol. 195, Oct. 2020, Art. no. 105568.

R. Kaur, H. GholamHosseini, and R. Sinha, “Synthetic images generation using conditional generative adversarial network for skin cancer classification,” in Proc. IEEE Region 10 Conf. (TENCON), Dec. 2021, pp. 381–386.

K. W. Lee and R. K. Y. Chin, “The effectiveness of data augmentation for melanoma skin cancer prediction using convolutional neural networks,” in Proc. IEEE 2nd Int. Conf. Artif. Intell. Eng. Technol. (IICAIET), Sep. 2020, pp. 1–6.