Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{181743,
        author = {PONNAM LALITHA and Dr.V JANARDHAN BABU},
        title = {Faster-CAN-FL: A Federated Convolutional Attention Network with AC-GAN for Privacy-Preserving Skin Cancer Classification},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {1},
        pages = {6041-6055},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=181743},
        abstract = {Early diagnosis and treatment of skin cancer depend heavily on its classification; nevertheless, manual classifi- cation is frequently laborious and prone to human error. Deep learning approaches have demonstrated significant promise in automating and enhancing the precision of skin lesion classi- fication, particularly in light of the swift progress of artificial intelligence. However, issues like data imbalance and restricted sample availability can sometimes hinder the efficacy of deep learning models. In order to improve model training, this study uses Auxiliary Classifier Generative Adversarial Networks (AC- GAN) to balance class distributions and create synthetic data. For the purpose of accurately and efficiently classifying skin cancer, we suggest a new hybrid model called Faster-CAN, which is a faster convolutional neural network architecture. Results from experiments show that adding data augmentation based on AC-GANs greatly enhances model performance. With a testing accuracy of 98.42%, precision of 98.50%, recall of 98.31%, and F1-score of 98.40%, the suggested Faster-CAN model produces exceptional results. These findings show how well synthetic data augmentation and a quicker, more efficient hybrid architecture work together to provide a reliable and scalable method for diagnosing skin cancer in clinical settings.},
        keywords = {Skin Cancer; Faster-CNN, Attention Mechanism, Federated Learning, HAM1000},
        month = {June},
        }