CNN-Based Skin Cancer Classification with Combined Max and Global Average Pooling

Chairani Fauzi; Fitra Salam S. Nagalay

doi:10.29207/resti.v9i6.6617

Authors

Chairani Fauzi Institut Informatika dan Bisnis Darmajaya
Fitra Salam S. Nagalay Institut Informatika dan Bisnis Darmajaya

DOI:

https://doi.org/10.29207/resti.v9i6.6617

Keywords:

convolutional neural network, global average pooling, max pooling, pooling, skin cancer

Abstract

Skin cancer is one of the most threatening diseases to human health, with an increase in new cases each year. Early detection plays a crucial role in improving recovery rates, however, conventional diagnostic methods such as biopsy are often invasive, time-consuming, and costly. To address this issue, artificial intelligence-based diagnostic systems, particularly Convolutional Neural Networks (CNNs), offer a promising solution for enhancing diagnostic accuracy and efficiency. This study aims to evaluate the performance of a CNN model that combines Max Pooling and Global Average Pooling (GAP) in detecting skin cancer from digital dermoscopic images. The ISIC (International Skin Imaging Collaboration) dataset was used, focusing on two classes: malignant and benign. The combination of Max Pooling and GAP is intended to increase model precision while reducing the risk of overfitting. The experimental results show that the proposed model achieved a precision of 96.35%, indicating strong performance in minimizing false positives. However, the recall was relatively low at 85.99%, suggesting reduced sensitivity in detecting malignant cases. The overall accuracy of the combined model was 91.68%, slightly lower than the Max Pooling-only model (91.79%). Although the combination does not significantly improve accuracy, it effectively enhances precision to 96.35%. This is a critical advantage in a clinical setting, as it directly translates to minimizing false positive diagnoses and preventing patients from undergoing unnecessary invasive procedures like biopsies.

Downloads

Download data is not yet available.

References

M. Duarte, S. S. Pedrosa, P. R. Khusial, and A. R. Madureira, “Exploring the interplay between stress mediators and skin microbiota in shaping age-related hallmarks: A review,” Mech. Ageing Dev., vol. 220, no. February, p. 111956, 2024, doi: 10.1016/j.mad.2024.111956.

E. R. Parker, J. Mo, and R. S. Goodman, “The dermatological manifestations of extreme weather events: A comprehensive review of skin disease and vulnerability,” J. Clim. Chang. Heal., vol. 8, no. July 2022, p. 100162, 2022, doi: 10.1016/j.joclim.2022.100162.

E. S. Nugroho, I. Ardiyanto, and H. A. Nugroho, “Systematic literature review of dermoscopic pigmented skin lesions classification using convolutional neural network (CNN),” Int. J. Adv. Intell. Informatics, vol. 9, no. 3, pp. 363–382, 2023, doi: 10.26555/ijain.v9i3.961.

International Agency for Research on Cancer (IARC), “Skin cancer,” IARC Newsletter, 2022. https://www.iarc.who.int/cancer-type/skin-cancer (accessed Oct. 29, 2024).

S. Taresia R, W. H. Suryawan, and A. W. Setiawan, “Early detection of skin cancer using K-NN and convolutional neural network,” J. Teknol. Inf. dan Ilmu Komput., vol. 7, no. 2, pp. 373–378, 2020, doi: 10.1142/9488.

A. Bassel, A. B. Abdulkareem, Z. A. A. Alyasseri, N. S. Sani, and H. J. Mohammed, “Automatic malignant and benign skin cancer classification using a hybrid deep learning approach,” Diagnostics, vol. 12, no. 10, 2022, doi: 10.3390/diagnostics12102472.

L. Hakim, Z. Sari, and Handhajani, “Classification of skin cancer pigment images using convolutional neural network,” J. Resti, vol. 1, no. 10, pp. 379–385, 2021, [Online]. Available: https://jurnal.iaii.or.id/index.php/RESTI/article/view/3001

K. Vayadande, A. A. Bhosle, R. G. Pawar, D. J. Joshi, P. A. Bailke, and O. Lohade, “Innovative approaches for skin disease identification in machine learning: A comprehensive study,” Oral Oncol. Reports, vol. 10, no. April, p. 100365, 2024, doi: 10.1016/j.oor.2024.100365.

A. F. Mavrogenis, P. Altsitzioglou, S. Tsukamoto, and C. Errani, “Biopsy techniques for musculoskeletal tumors: basic principles and specialized techniques,” Curr. Oncol., vol. 31, no. 2, pp. 900–917, 2024, doi: 10.3390/curroncol31020067.

S. Rajarajeswari, J. Prassanna, M. Abdul Quadir, J. Christy Jackson, S. Sharma, and B. Rajesh, “Skin cancer detection using deep learning,” Res. J. Pharm. Technol., vol. 15, no. 10, pp. 4519–4525, 2022, doi: 10.52711/0974-360X.2022.00758.

J. Triloka, H. Hartono, and S. Sutedi, “Detection of SQL injection attack using machine learning based on natural language processing,” Int. J. Artif. Intell. Res., vol. 6, no. 2, 2022, doi: 10.29099/ijair.v6i2.355.

M. A. Saputra, Reyhan Adi Putra, Davito Rasendriya Rizqullah Asyrofi, “Implementation of convolutional neural network (CNN) to detect mask usage in images,” J. Inform. dan Tek. Elektro Terap., vol. 11, no. 3, pp. 710–714, 2023, doi: 10.23960/jitet.v11i3.3286.

A. Zafar et al., “A Comparison of Pooling methods for convolutional neural networks,” Appl. Sci., vol. 12, no. 17, pp. 1–21, 2022, doi: 10.3390/app12178643.

J. J. Liu, Q. Hou, Z. A. Liu, and M. M. Cheng, “PoolNet+: exploring the potential of pooling for salient object detection,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 45, no. 1, pp. 887–904, 2023, doi: 10.1109/TPAMI.2021.3140168.

R. L. Kumar, J. Kakarla, B. V. Isunuri, and M. Singh, “Multi-class brain tumor classification using residual network and global average pooling,” Multimed. Tools Appl., vol. 80, no. 9, pp. 13429–13438, 2021, doi: 10.1007/s11042-020-10335-4.

R. R. Saputro, A. Junaidi, and W. A. Saputra, “Classification of skin cancer disease using convolutional neural network method (case study: melanoma),” J. Dinda Data Sci. Inf. Technol. Data Anal., vol. 2, no. 1, pp. 52–57, 2022, doi: 10.20895/dinda.v2i1.349.

A. Nawaz, Khadija Zanib, Atika Shabir, Iqra Li, Jianqiang Wang, Yu Mahmood, Tariq Rehman, “Skin cancer detection using dermoscopic images with convolutional neural network,” Sci. Rep., pp. 254–258, 2025, doi: 10.1038/s41598-025-91446-6.

M. Saini and S. Susan, “Tackling class imbalance in computer vision: a contemporary review,” Artif. Intell. Rev., vol. 56, no. July, pp. 1279–1335, 2023, doi: 10.1007/s10462-023-10557-6.

J. Wang et al., “Generalizing to unseen domains: a survey on domain generalization,” IEEE Trans. Knowl. Data Eng., vol. 35, no. 8, pp. 8052–8072, 2023, doi: 10.1109/TKDE.2022.3178128.

S. Y. Irianto, R. Yunandar, M. S. Hasibuan, D. A. Dewi, and N. Pitsachart, “Early identification of skin cancer using region growing technique and a deep learning algorithm,” HighTech Innov. J., vol. 5, no. 3, pp. 640–662, 2024, doi: 10.28991/HIJ-2024-05-03-07.

E. Academy, “Mengapa perlu menormalkan nilai piksel sebelum melatih model?,” EITCA Academy, 2023. https://id.eitca.org/kecerdasan-buatan/eitc-ai-tf-tensorflow-fundamental/aliran-tensor-js/menggunakan-tensorflow-untuk-mengklasifikasikan-gambar-pakaian/ulasan-pemeriksaan-tensorflow-js-mengapa-perlu-menormalkan-nilai-piksel-sebelum-melatih-model/ (accessed Nov. 21, 2024).

W. Bakx, B. Gorte, and K. Grabmaier, The core of GIScience: A process-based approach. ITC, Faculty of geo-information Science and Earth Observation, 2011.

E. Beyazit, J. Kozaczuk, B. Li, V. Wallace, and B. Fadlallah Amazon, “An inductive bias for tabular deep learning,” Conf. Neural Inf. Process. Syst., no. NeurIPS, 2023.

I. Pacal, “A novel Swin transformer approach utilizing residual multi-layer perceptron for diagnosing brain tumors in MRI images,” Int. J. Mach. Learn. Cybern., vol. 15, no. 9, pp. 3579–3597, 2024, doi: 10.1007/s13042-024-02110-w.

G. P. H. P. Gusti, E. Haerani, F. Syafria, F. Yanto, and S. K. Gusti, “Implementation of convolutional neural network algorithm (ResNet-50) for benign and malignant skin cancer classification,” MALCOM Indones. J. Mach. Learn. Comput. Sci., vol. 4, no. 3, pp. 984–992, 2024, doi: 10.57152/malcom.v4i3.1398.

E. Setia Budi, A. Nofriyaldi Chan, P. Priscillia Alda, and M. Arif Fauzi Idris, “Optimization of machine learning model for image classification and prediction using convolutional neural network algorithm,” Media Online, vol. 4, no. 5, p. 509, 2024, [Online]. Available: https://djournals.com/resolusi

F. Rauf et al., “Artificial intelligence assisted common maternal fetal planes prediction from ultrasound images based on information fusion of customized convolutional neural networks,” Front. Med., vol. 11, no. October, 2024, doi: 10.3389/fmed.2024.1486995.

X. Yang, S. Yu, and W. Xu, “Enhanced convolutional neural networks for improved image classification,” 2025, [Online]. Available: http://arxiv.org/abs/2502.00663

B. Faye, M. Lebbah, and H. Azzag, “Supervised batch normalization,” 2024, [Online]. Available: http://arxiv.org/abs/2405.17027

L. Zhao and Z. Zhang, “A improved pooling method for convolutional neural networks,” Sci. Rep., vol. 14, no. 1, pp. 1–23, 2024, doi: 10.1038/s41598-024-51258-6.

X. Kong, X. Liu, J. Gu, Y. Qiao, and C. Dong, “Reflash dropout in image super-resolution,” Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit., vol. 2022-June, pp. 5992–6002, 2022, doi: 10.1109/CVPR52688.2022.00591.

M. M. Taye, “Theoretical understanding of convolutional neural network: concepts, architectures, applications, future directions,” Computation, vol. 11, no. 3, 2023, doi: 10.3390/computation11030052.

M. Lin, Q. Chen, and S. Yan, “Network in network,” 2nd Int. Conf. Learn. Represent. ICLR 2014 - Conf. Track Proc., pp. 1–10, 2014.

Sutedi, M. Royan, F. Maulana, M. Agarina, and A. Suryadi, “Brain tumor detection on magnetic resonance imaging using deep neural network,” vol. 7, no. 1, 2023.

L. A. Yates, Z. Aandahl, S. A. Richards, and B. W. Brook, “Cross validation for model selection: A review with examples from ecology,” Ecol. Monogr., vol. 93, no. 1, pp. 1–24, 2023, doi: 10.1002/ecm.1557.

V. M. Patro and M. R. Patra, “A novel approach to compute confusion matrix for classification of n-class attributes with feature selection,” Trans. Mach. Learn. Artif. Intell., vol. 3, no. 2, 2015.

L. Raditya, “Metrik evaluasi untuk klasifikasi,” Medium, 2022. https://luthfiradity.medium.com/evaluating-classification-model-a2e19a50acee (accessed Feb. 02, 2025).

K. S. Nugroho, “Confusion matrix untuk evaluasi model pada supervised learning,” Medium, 2019.

M. E. Klontzas, G. Kalarakis, E. Koltsakis, T. Papathomas, A. H. Karantanas, and A. Tzortzakakis, “Convolutional neural networks for the differentiation between benign and malignant renal tumors with a multicenter international computed tomography dataset,” Insights Imaging, vol. 15, no. 1, 2024, doi: 10.1186/s13244-023-01601-8.

K. Faryna, J. van der Laak, and G. Litjens, “Automatic data augmentation to improve generalization of deep learning in H&E stained histopathology,” Comput. Biol. Med., vol. 170, no. October 2023, p. 108018, 2024, doi: 10.1016/j.compbiomed.2024.108018.

E. Goceri, Medical image data augmentation: techniques, comparisons and interpretations, vol. 56, no. 11. Springer Netherlands, 2023. doi: 10.1007/s10462-023-10453-z.

O. Rainio and R. Klén, “Comparison of simple augmentation transformations for a convolutional neural network classifying medical images,” Signal, Image Video Process., vol. 18, no. 4, pp. 3353–3360, 2024, doi: 10.1007/s11760-024-02998-5.

L. Huang, J. Qin, Y. Zhou, F. Zhu, L. Liu, and L. Shao, “Normalization techniques in training DNNs: methodology, analysis and application,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 45, no. 8, pp. 10173–10196, 2023, doi: 10.1109/TPAMI.2023.3250241.

C. Ozdemir, Y. Dogan, and Y. Kaya, “A new local pooling approach for convolutional neural network: local binary pattern,” Multimed. Tools Appl., vol. 83, no. 12, pp. 34137–34151, 2024, doi: 10.1007/s11042-023-17540-x.

Z. Liu, Z. Xu, J. Jin, Z. Shen, and T. Darrell, “Dropout reduces underfitting,” Proc. Mach. Learn. Res., vol. 202, pp. 21715–21729, 2023.

D. T. Tran, N. Shimada, and J. H. Lee, “Triple-sigmoid activation function for deep open-set recognition,” IEEE Access, vol. 10, no. June, pp. 77668–77678, 2022, doi: 10.1109/ACCESS.2022.3192621.

GeeksforGeeks, “Why an Increasing Validation Loss and Validation Accuracy Signifies Overfitting?,” GeeksforGeeks, 2023. https://www.geeksforgeeks.org/why-an-increasing-validation-loss-and-validation-accuracy-signifies-overfitting (accessed Feb. 02, 2025).

Lutz Prechelt, “Early stopping - but when?,” no. March 2000, 2015, doi: 10.1007/3-540-49430-8.

M. Valdenegro-Toro and M. Sabatelli, “Machine learning students overfit to overfitting,” Proc. Mach. Learn. Res., vol. 207, pp. 46–51, 2022.

X. Ying, “An overview of overfitting and its solutions,” J. Phys. Conf. Ser., vol. 1168, no. 2, 2019, doi: 10.1088/1742-6596/1168/2/022022.