An Interpretable Deep Learning Approach for Skin Cancer Categorization

This repository contains the code and datasets used in the paper titled "An Interpretable Deep Learning Approach for Skin Cancer Categorization" accepted and presented at the 26th International Conference on Computer and Information Technology (ICCIT) 2023.

Paper Link: PDF

Table of Contents

• Dataset
• Result
• Citation

Dataset

We used in this paper publicly available HAM10000 Dataset

Result

Model-specific Classification Report of Weighted Average

Models	Accuracy	Precision	Recall	F1 Score
XceptionNet	88.72%	0.89	0.89	0.89
EfficientNetV2S	88.02%	0.88	0.88	0.88
InceptionResNetV2	85.73%	0.86	0.86	0.85
EfficientNetV2M	85.02%	0.89	0.89	0.89

Citation

If you found this code helpful please consider citing,

@INPROCEEDINGS{10508527,
            author={Mahmud, Faysal and Mahfiz, Md. Mahin and Kabir, Md. Zobayer Ibna and Abdullah, Yusha},
            booktitle={2023 26th International Conference on Computer and Information Technology (ICCIT)}, 
            title={An Interpretable Deep Learning Approach for Skin Cancer Categorization}, 
            year={2023},
            volume={},
            number={},
            pages={1-6},
            keywords={Deep learning;Visualization;Explainable AI;Computational modeling;Medical services;Skin;Lesions;Skin Cancer Detection;Deep Learning;Pre-trained Models;Convolutional             Neural Networks (CNN);HAM10000;Medical Imaging;Explainable Artificial Intelligence (XAI)},
            doi={10.1109/ICCIT60459.2023.10508527}
}

License

This repository is licensed under the MIT License. See the LICENSE file for more information.