Authors : Arathi V. ; Harshitha N. C.; Kruthi K. Shetty; Chayadevi M. L.

Volume/Issue : Volume 11 - 2026, Issue 9 - September

Google Scholar : https://tinyurl.com/38es6a7b

DOI : https://doi.org/10.38124/ijisrt/26sep901

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Abstract : Automated melanoma detection using deep learning has reached dermatologist-level accuracy on benchmark datasets, yet its clinical deployment remains constrained by a well-documented weakness: performance degrades sharply on darker skin tones because public dermoscopic archives are dominated by lighter-skinned patients. This paper presents a bias-aware, explainable framework for melanoma classification that treats data quality, subgroup fairness, and interpretability as first-class design objectives rather than post-hoc checks. Dermoscopic images drawn from the ISIC archive and the Diverse Dermatology Images (DDI) dataset are first cleaned using perceptual-hash-based duplicate removal and DullRazor-based hair/artifact suppression, then passed through a ResNet50 backbone fine-tuned via transfer learning for binary melanoma-versus-benign classification.

Keywords : Melanoma Detection; Algorithmic Fairness; Deep Learning; ResNet50; Grad-CAM; Skin Tone Bias; Dermoscopic Imaging; Explainable AI.

References :

1. A. Esteva, B. Kuprel, R. A. Novoa, J. Ko, S. M. Swetter, H. M. Blau, and S. Thrun, "Dermatologist-level classification of skin cancer with deep neural networks," Nature, vol. 542, no. 7639, pp. 115–118, 2017.
2. R. Daneshjou, M. Smith, M. Sun, et al., "Lack of transparency and potential bias in artificial intelligence data sets and algorithms: A scoping review in dermatology," npj Digital Medicine, vol. 6, pp. 1–10, 2023.
3. J. Gichoya, I. Banerjee, A. Bhimireddy, et al., "AI recognition of patient race in medical imaging: A modelling study," Nature Medicine, vol. 29, pp. 144–150, 2023.
4. M. Groh, C. Harris, R. Soenksen, et al., "Evaluating deep neural networks trained on clinical images in dermatology with diverse skin tones," Proceedings of the National Academy of Sciences (PNAS), vol. 121, no. 5, e2206834121, 2024.
5. H. Liu, L. Wang, Y. Chen, et al., "Skin lesion classification using deep learning: A comprehensive review," IEEE Access, vol. 12, pp. 45678–45695, 2024.
6. T. Han, J. Lee, S. Park, et al., "Improving skin color diversity in skin cancer detection using data augmentation techniques," IEEE Transactions on Medical Imaging, vol. 43, no. 2, pp. 789–801, 2024.
7. Y. Li, X. Zhang, and R. Kumar, "Bias and fairness in deep learning-based melanoma detection systems," Artificial Intelligence in Medicine, vol. 150, 102345, 2025.
8. K. Brinker, A. Hekler, J. S. Enk, et al., "Deep learning outperformed 136 dermatologists in a head-to-head dermoscopic melanoma image classification task," European Journal of Cancer, vol. 113, pp. 47–54, 2023.
9. N. Codella, D. Gutman, M. Celebi, et al., "Skin lesion analysis toward melanoma detection: A challenge at the International Skin Imaging Collaboration (ISIC)," IEEE Journal of Biomedical and Health Informatics, vol. 23, no. 2, pp. 547–552, 2023.
10. P. Tschandl, C. Rosendahl, and H. Kittler, "Human–computer collaboration for skin cancer recognition," Nature Medicine, vol. 26, pp. 1229–1234, 2024.