/An-Interpretable-Deep-Learning-Approach-for-Skin-Cancer-Categorization-IEEE2023

Multiclass skin cancer detection using explainable AI for checking the models' robustness

Primary LanguageJupyter NotebookMIT LicenseMIT

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

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}
}