TY  - JOUR
AU  - Negi, Simran
AU  - Joshi, Aditya
PY  - 2025
DA  - 2025/12/31
TI  - Transformer Enabled ResNet Based Automated Skin Cancer Detection System
JO  - Biomedical Informatics and Smart Healthcare
T2  - Biomedical Informatics and Smart Healthcare
JF  - Biomedical Informatics and Smart Healthcare
VL  - 1
IS  - 3
SP  - 149
EP  - 154
DO  - 10.62762/BISH.2025.444143
UR  - https://www.icck.org/article/abs/BISH.2025.444143
KW  - ResNet50
KW  - deep learning
KW  - convolutional neural networks
KW  - skin cancer
AB  - Early diagnosis plays a critical role in the successful skin cancer treatment. To classify skin lesions as benign or malignant, this study suggested a deep learning-enabled method that combines a Transformer module with ResNet50. At first ResNet50, a powerful convolutional neural network, to extract the image features and then enhance the model with a Transformer layer to improve the model accuracy is used. The final model is fine-tuned to achieve better accuracy. This approach shows improved classification results compared to the traditional model. The results signify that combining CNN-based feature extraction with Transformer-enabled global attention suggestively improves skin lesion classification, making the suggested framework a promising method for early and consistent skin cancer detection in intelligent healthcare systems.
SN  - 3068-5524
PB  - Institute of Central Computation and Knowledge
LA  - English
ER  - 

@article{Negi2025Transforme,
  author = {Simran Negi and Aditya Joshi},
  title = {Transformer Enabled ResNet Based Automated Skin Cancer Detection System},
  journal = {Biomedical Informatics and Smart Healthcare},
  year = {2025},
  volume = {1},
  number = {3},
  pages = {149-154},
  doi = {10.62762/BISH.2025.444143},
  url = {https://www.icck.org/article/abs/BISH.2025.444143},
  abstract = {Early diagnosis plays a critical role in the successful skin cancer treatment. To classify skin lesions as benign or malignant, this study suggested a deep learning-enabled method that combines a Transformer module with ResNet50. At first ResNet50, a powerful convolutional neural network, to extract the image features and then enhance the model with a Transformer layer to improve the model accuracy is used. The final model is fine-tuned to achieve better accuracy. This approach shows improved classification results compared to the traditional model. The results signify that combining CNN-based feature extraction with Transformer-enabled global attention suggestively improves skin lesion classification, making the suggested framework a promising method for early and consistent skin cancer detection in intelligent healthcare systems.},
  keywords = {ResNet50, deep learning, convolutional neural networks, skin cancer},
  issn = {3068-5524},
  publisher = {Institute of Central Computation and Knowledge}
}