Citation :

Priyadarshini Mayigowda, Zafar Ali Khan N, "Tone Aware Convolutional Neural Network with Conditional Feature Pyramid Network for Skin Cancer Classification," International Journal of Engineering Trends and Technology (IJETT), vol. 74, no. 4, pp. 342-359, 2026. Crossref, https://doi.org/10.14445/22315381/IJETT-V74I4P126

Abstract

Skin cancer classification involves identifying and categorizing skin lesions as either malignant or benign, depending on visual appearance, which helps doctors detect and monitor skin cancer more effectively. However, classifying the severity of skin cancer from dermoscopic images is difficult due to subtle and overlapping visual features, which leads to inconsistencies and enhances misclassification. This research proposes a Tone Aware Convolutional Neural Network with Conditional Feature Pyramid Network (TACNN-CFPN) to classify skin cancer accurately. In TACNN, CFPN is incorporated, which enables the model to adaptively perform specified feature extraction paths based on skin tone, become more sensitive to the visual characteristics of each skin tone, and lead to more accurate identification of lesions. TACNN adapts feature learning to the specific skin tone of input, which addresses visual variations in lesion appearance and enhances classification accuracy across diverse populations. Resizing ensures uniform image dimension, whereas over- and undersampling methods balance class distribution that minimizes bias and enhances model generalization. Therefore, the TACNN-CFPN achieves better accuracy of 98.84%, 99.84%, 98.87%, and 98.53% on the ISIC-2019, ISIC-2020, HAM10000, and PAU-UFES-20 datasets compared to existing methods, namely Deep CNN.

Keywords

Conditional Feature Pyramid Network, Tone Aware Convolutional Neural Network, Resizing, Skin Cancer, Undersampling.

References

[1] Mawaddah Harahap et al., “Skin Cancer Classification using EfficientNet Architecture,” Bulletin of Electrical Engineering and Informatics, vol. 13, no. 4, pp. 2716-2728, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[2] Neven Saleh, Mohammed A. Hassan, and Ahmed M. Salaheldin, “Skin Cancer Classification based on an Optimized Convolutional Neural Network and Multicriteria Decision-Making,” Scientific Reports, vol. 14, no. 1, pp. 1-19, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[3] Usha Thirugnanam, and Nalini Joseph, “Skin Cancer Classification using Premature Convergence Strategy-based Artificial Jelly Search Optimization with Convolutional Neural Network,” International Journal of Intelligent Engineering & Systems, vol. 17, no. 5, pp. 848-858, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[4] Galib Muhammad Shahriar Himel et al., “Skin Cancer Segmentation and Classification using Vision Transformer for Automatic Analysis in Dermatoscopy‐based Noninvasive Digital System,” International Journal of Biomedical Imaging, vol. 2024, pp. 1-18, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[5] Günay İlker, and İnik Özkan, “SADASNet: A Selective and Adaptive Deep Architecture Search Network with Hyperparameter Optimization for Robust Skin Cancer Classification,” Diagnostics, vol. 15, no. 5, pp. 1-28, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[6] Ali Atshan Abdulredah et al., “Towards Unbiased Skin Cancer Classification using Deep Feature Fusion,” BMC Medical Informatics and Decision Making, vol. 25, no. 1, pp. 1-22, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[7] Sara M.M. Abohashish, Hanan H. Amin, and E.I. Elsedimy, “Enhanced Melanoma and Non-Melanoma Skin Cancer Classification using a Hybrid LSTM-CNN Model,” Scientific Reports, vol. 15, no. 1, pp. 1-23, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[8] J.D. Dorathi Jayaseeli et al., “An Intelligent Framework for Skin Cancer Detection and Classification using Fusion of Squeeze-Excitation-Densenet with Metaheuristic-Driven Ensemble Deep Learning Models,” Scientific Reports, vol. 15, no. 1, pp. 1-23, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[9] Umesh Kumar Lilhore et al., “A Precise Model for Skin Cancer Diagnosis using Hybrid U-Net and Improved MobileNet-V3 with Hyperparameters Optimization,” Scientific Reports, vol. 14, no. 1, pp. 1-24, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[10] K.M. Monica et al., “Melanoma Skin Cancer Detection using Mask-RCNN with Modified GRU Model,” Frontiers in Physiology, vol. 14, pp. 1-12, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[11] Ishak Pacal, Melek Alaftekin, and Ferhat Devrim Zengul, “Enhancing Skin Cancer Diagnosis using Swin Transformer with Hybrid Shifted Window-based Multi-Head Self-Attention and SwiGLU-based MLP,” Journal of Imaging Informatics in Medicine, vol. 37, no. 6, pp. 3174-3192, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[12] Abayomi Bello, Sin-Chun Ng, and Man-Fai Leung, “Skin Cancer Classification using Fine-Tuned Transfer Learning of DENSENET-121,” Applied Sciences, vol. 14, no. 17, pp. 1-25, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[13] Mohamed Hosny et al., “Attention-based Convolutional Neural Network Model for Skin Cancer Classification,” IEEE Access, vol. 13, pp. 172027-172050, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[14] Rodrigue Bogne Tchema, Anastasis C. Polycarpou, and Marios Nestoros, “Skin Cancer Classification using Machine Learning,” Multimedia Tools and Applications, vol. 84, no. 6, pp. 3239-325, 2025.
[CrossRef] [Publisher Link]

[15] Vasuja Devi Midasala et al., “MFEUsLNet: Skin Cancer Detection and Classification using Integrated AI with Multilevel Feature Extraction-based Unsupervised Learning,” Engineering Science and Technology, an International Journal, vol. 51, pp. 1-17, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[16] Burhanettin Ozdemir, and Ishak Pacal, “A Robust Deep Learning Framework for Multiclass Skin Cancer Classification,” Scientific Reports, vol. 15, no. 1, pp. 1-19, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[17] Rizwan Ali et al., “A Novel SpaSA-based Hyperparameter-Optimized FCEDN with Adaptive CNN Classification for Skin Cancer Detection,” Scientific Reports, vol. 14, no. 1, pp. 1-17, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[18] Hugo Vega-Huerta et al., “Classification Model of Skin Cancer using Convolutional Neural Network,” Information Systems Engineering, vol. 30, no. 2, pp. 387-394, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[19] Ahmet Nusret Toprak, and Ibrahim Aruk, “A Hybrid Convolutional Neural Network Model for the Classification of Multi‐Class Skin Cancer,” International Journal of Imaging Systems and Technology, vol. 34, no. 5, pp. 1-18, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[20] Fallah H. Najjar et al., “Transformer-Aided Skin Cancer Classification using VGG19-based Feature Encoding,” Scientific Reports, vol. 15, no. 1, pp. 1-18, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[21] Essam H. Houssein et al., “An Effective Multiclass Skin Cancer Classification Approach based on Deep Convolutional Neural Network,” Cluster Computing, vol. 27, no. 9, pp. 12799-12819, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[22] Zhanlin Ji et al., “EFAM-Net: A Multi-Class Skin Lesion Classification Model Utilizing Enhanced Feature Fusion and Attention Mechanisms,” IEEE Access, vol. 12, pp. 143029-143041, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[23] Yousef S. Alsahafi, Mohamed A. Kassem, and Khalid M. Hosny, “Skin-Net: A Novel Deep Residual Network for Skin Lesions Classification using Multilevel Feature Extraction and Cross-Channel Correlation with Detection of Outlier,” Journal of Big Data, vol. 10, no. 1, pp. 1-23, 2023.
[CrossRef] [Google Scholar] [Publisher Link]

[24] Sasmita Padhy et al., “Temporal Integration of ResNet Features with LSTM for Enhanced Skin Lesion Classification,” Results in Engineering, vol. 25, pp. 1-13, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[25] V.S.S. Bala Tripura Sathvika et al., “Pipelined Structure in the Classification of Skin Lesions based on AlexNet CNN and SVM Model with Bi-Sectional Texture Features,” IEEE Access, vol. 12, pp. 57366-57380, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[26] Sara Medhat et al., “Iterative Magnitude Pruning-based Light Version of AlexNet for Skin Cancer Classification,” Neural Computing and Applications, vol. 36, no. 3, pp. 1413-1428, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[27] Amany M. Sarhan et al., “Achieving High-Accuracy Skin Cancer Classification with Deep Learning Optimized by Ant Colony Algorithm,” Journal of Electrical Systems and Information Technology, vol. 12, no. 1, pp. 1-25, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[28] Omneya Attallah, “Skin Cancer Classification Leveraging Multi-Directional Compact Convolutional Neural Network Ensembles and Gabor Wavelets,” Scientific Reports, vol. 14, no. 1, pp. 1-19, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[29] Geunho Jung, Semin Kim, and Sangwook Yoo, “Skin Tone Analysis through Skin Tone Map Generation with Optical Approach and Deep Learning,” Skin Research and Technology, vol. 30, no. 10, pp. 1-8, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[30] K Scott Mader, Skin Cancer MNIST: HAM10000, 2018. [Online]. Available: https://www.kaggle.com/datasets/kmader/skin-cancer-mnist-ham10000

[31] Prasad Maharana, ISIC 2019 Skin Lesion images for classification, 2018. [Online]. Available: https://www.kaggle.com/datasets/salviohexia/isic-2019-skin-lesion-images-for-classification  

[32] Sumaiya Binte Shahid, ISIC Challenge Dataset-2020, 2020. [Online]. Available: https://www.kaggle.com/datasets/sumaiyabinteshahid/isic-challenge-dataset-2020

[33] Mahdavi, Skin Cancer (PAD-UFES-20), 2020. [Online]. Available:  https://www.kaggle.com/datasets/mahdavi1202/skin-cancer

[34] Nasraldeen Alnor Adam Khleel, and Károly Nehéz, “Software Defect Prediction using a Bidirectional LSTM Network Combined with Oversampling Techniques,” Cluster Computing, vol. 27, no. 3, pp. 3615-3638, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[35] D.R. Manjunath et al., “Predicting Diabetic Retinopathy and Nephropathy Complications using Machine Learning Techniques,” IEEE Access, vol. 13, pp. 70228-70253, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[36] Ahmed Hatem Soudy et al., “Deepfake Detection using Convolutional Vision Transformers and Convolutional Neural Networks,” Neural Computing and Applications, vol. 36, no. 31, pp. 19759-19775, 2024.
[CrossRef] [Google Scholar] [Publisher Link]