Skin Lesion Classification: A Deep Learning Approach with Local Interpretable Model-Agnostic Explanations (LIME) for Explainable Artificial Intelligence (XAI)

Sin Yi Hong - Tunku Abdul Rahman University of Management and Technology, Kuala Lumpur, Malaysia
Lih Poh Lin - Tunku Abdul Rahman University of Management and Technology, Kuala Lumpur, Malaysia


Citation Format:



DOI: http://dx.doi.org/10.62527/joiv.8.3-2.3022

Abstract


The classification of skin cancer is crucial as the chance of survival increases significantly with timely and accurate treatment. Convolution Neural Networks (CNNs) have proven effective in classifying skin cancer. However, CNN models are often regarded as "black boxes”, due to the lack of transparency in the decision-making. Therefore, explainable artificial intelligence (XAI) has emerged as a tool for understanding AI decisions. This study employed a CNN model, VGG16, to classify five skin lesion classes. The hyperparameters were adjusted to optimize its classification performance. The best hyperparameter settings were 50 epochs, a 0.1 dropout rate, and the Adam optimizer with a 0.001 learning rate. The VGG16 model demonstrated satisfactory classification performance. The Local Interpretable Model-Agnostic Explanations (LIME) method was implemented as the XAI tool to justify the predictions made by VGG16. The LIME explanation revealed that the correct predictions made by VGG16 were owing to its truthful extraction of the cancer or lesion area, especially for the “vascular lesion” class. Meanwhile, inaccurate classifications were attributed to VGG16 extraction of the background and insignificant parts of the skin as core features. In conclusion, The LIME model allowed visual inspection of the features selected by VGG16, paving the way for improving the CNN model for better feature extraction and classification of skin lesions, offering a promising direction for future research. 


Keywords


CNN; Deep Learning; Explainable AI; Skin Cancer; Local Interpretable Model-Agnostic Explanations; VGG16; XAI

Full Text:

PDF

References


C. M. Relvas, S. G. Santos, M. J. Oliveira, F. D. Magalhães, and A. M. Pinto, “Nanomaterials for Skin Cancer Photoimmunotherapy,” Biomedicines, vol. 11, no. 5, 2023, doi:10.3390/biomedicines11051292.

V. Gruber et al., “Common Benign Melanocytic and Non-Melanocytic Skin Tumors among the Elderly: Results of the Graz Study on Health and Aging,” Dermatology, vol. 239, no. 3, pp. 379–386, 2023, doi: 10.1159/000529219.

M. Tahir, A. Naeem, H. Malik, J. Tanveer, R. A. Naqvi, and S. W. Lee, “DSCC_Net: Multi-Classification Deep Learning Models for Diagnosing of Skin Cancer Using Dermoscopic Images,” Cancers (Basel)., vol. 15, no. 7, 2023, doi: 10.3390/cancers15072179.

R. L. Siegel, K. D. Miller, N. S. Wagle, and A. Jemal, “Cancer statistics, 2023,” CA. Cancer J. Clin., vol. 73, no. 1, pp. 17–48, 2023, doi: 10.3322/caac.21763.

A. Esteva et al., “Dermatologist-level classification of skin cancer with deep neural networks,” Nature, vol. 542, no. 7639, pp. 115–118, 2017, doi: 10.1038/nature21056.

Y. Wu, B. Chen, A. Zeng, D. Pan, R. Wang, and S. Zhao, “Skin Cancer Classification With Deep Learning: A Systematic Review,” Front. Oncol., vol. 12, pp. 1–20, 2022, doi:10.3389/fonc.2022.893972.

A. S. Jaradat et al., “Automated Monkeypox Skin Lesion Detection Using Deep Learning and Transfer Learning Techniques,” Int. J. Environ. Res. Public Health, vol. 20, no. 5, 2023, doi:10.3390/ijerph20054422.

Y. Nie, P. Sommella, M. Carratù, M. O’Nils, and J. Lundgren, “A Deep CNN Transformer Hybrid Model for Skin Lesion Classification of Dermoscopic Images Using Focal Loss,” Diagnostics, vol. 13, no. 1, pp. 1–18, 2023, doi: 10.3390/diagnostics13010072.

K. M. Hosny and M. A. Kassem, “Refined Residual Deep Convolutional Network for Skin Lesion Classification,” J. Digit. Imaging, vol. 35, no. 2, pp. 258–280, 2022, doi: 10.1007/s10278-021-00552-0.

A. Mahbod, G. Schaefer, C. Wang, R. Ecker, and I. Ellinger, “Skin lesion classification using hybrid deep neural networks,” 2019 IEEE Int. Conf. Acoust. Speech Signal Process., pp. 1229–1233, 2019, doi:10.1109/icassp.2019.8683352.

A. Khamparia, P. K. Singh, P. Rani, D. Samanta, A. Khanna, and B. Bhushan, “An internet of health things-driven deep learning framework for detection and classification of skin cancer using transfer learning,” Trans. Emerg. Telecommun. Technol., vol. 32, no. 7, pp. 1–11, 2021, doi: 10.1002/ett.3963.

Z. Wu et al., “Studies on Different CNN Algorithms for Face Skin Disease Classification Based on Clinical Images,” IEEE Access, vol. 7, pp. 66505–66511, 2019, doi: 10.1109/ACCESS.2019.2918221.

C. Linse, H. Alshazly, and T. Martinetz, “A walk in the black-box: 3D visualization of large neural networks in virtual reality,” Neural Comput. Appl., vol. 34, no. 23, pp. 21237–21252, 2022, doi:10.1007/s00521-022-07608-4.

Y. Zhang, Y. Weng, and J. Lund, “Applications of Explainable Artificial Intelligence in Diagnosis and Surgery,” Diagnostics, vol. 12, no. 2, 2022, doi: 10.3390/diagnostics12020237.

K. Borys et al., “Explainable AI in medical imaging: An overview for clinical practitioners - Saliency-based XAI approaches,” Eur. J. Radiol., vol. 162, p. 110787, 2023, doi: 10.1016/j.ejrad.2023.110787.

H. Kaur, S. Jindal, and R. Manduchi, “Rethinking Model-Based Gaze Estimation,” Proc ACM Comput Graph Interact Tech, vol. 5, no. 2, pp. 1–28, 2022, doi: 10.1145/3530797.

K. Young, G. Booth, B. Simpson, R. Dutton, and S. Sally, “Deep Neural Network or Dermatologist?,” in Interpretability of Machine Intelligence in Medical Image Computing and Multimodal Learning for Clinical Decision Support, 2019, pp. 48–55. doi: 10.1007/978-3-030-33850-3_6.

J. H. Ong, K. M. Goh, and L. L. Lim, “Comparative Analysis of Explainable Artificial Intelligence for COVID-19 Diagnosis on CXR Image,” Proc. 2021 IEEE Int. Conf. Signal Image Process. Appl. ICSIPA 2021, pp. 185–190, 2021, doi:10.1109/icsipa52582.2021.9576766.

M. Xiao, L. Zhang, W. Shi, J. Liu, W. He, and Z. Jiang, “A visualization method based on the Grad-CAM for medical image segmentation model,” 2021 Int. Conf. Electron. Inf. Eng. Comput. Sci. EIECS 2021, pp. 242–247, 2021, doi:10.1109/eiecs53707.2021.9587953.

J. Rebane, I. Samsten, P. Pantelidis, and P. Papapetrou, “Assessing the clinical validity of attention-based and SHAP temporal explanations for adverse drug event predictions,” Proc. - IEEE Symp. Comput. Med. Syst., pp. 235–240, 2021, doi:10.1109/cbms52027.2021.00025.

H. T. T. Nguyen, Q. C. Hung, K. V. T. Nguyen, and D. K. P. Nguyen, “Evaluation of explainable artificial intelligence: Shap, lime, and cam,” Proc. FPT AI Conf., pp. 1–6, 2022, doi: 10.30727/0235-1188-2022-65-1-72-90.

A. Katanskiy, “Skin Cancer ISIC,” Kaggle. [Online]. Available: https://www.kaggle.com/datasets/nodoubttome/skin-cancer9-classesisic

I. The International Skin Imaging Collaboration, “ISIC Challenge Datasets.” [Online]. Available: https://challenge.isic-archive.com/data/#2019

P. Tschandl, C. Rosendahl, and H. Kittler, “Data Descriptor : The HAM 10000 dataset , a large collection of multi-source dermatoscopic images of common pigmented skin lesions,” Nat. Publ. Gr., pp. 1–9, 2018, doi: 10.1038/sdata.2018.161.

N. C. F. Codella et al., “Skin lesion analysis toward melanoma detection: a challenge at the 2017 international symposium on biomedical imaging (ISBI), hosted by the international skin imaging collaboration (ISIC),” Proceeding 2018 IEEE 15th Int. Symp. Biomed. Imaging, pp. 168–172, 2018, doi: 10.1109/ISBI.2018.8363547.

M. Combalia et al., “BCN20000: Dermoscopic Lesions in the Wild", 2019.” [Online]. Available: https://arxiv.org/abs/1908.02288

K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale image recognition,” 3rd Int. Conf. Learn. Represent. ICLR 2015 - Conf. Track Proc., pp. 1–14, 2015, doi:10.48550/arXiv.1409.1556.

Z. Liu et al., “Improved Kiwifruit Detection Using Pre-Trained VGG16 with RGB and NIR Information Fusion,” IEEE Access, vol. 8, pp. 2327–2336, 2020, doi: 10.1109/ACCESS.2019.2962513.

C. Yang and H. Wei, “Plant Species Recognition Using Triangle-Distance Representation,” IEEE Access, vol. 7, pp. 178108–178120, 2019, doi: 10.1109/access.2019.2958416.

S. Ioffe and C. Szegedy, “Batch normalization: Accelerating deep network training by reducing internal covariate shift,” 32nd Int. Conf. Mach. Learn. ICML 2015, vol. 1, pp. 448–456, 2015, doi:10.48550/arXiv.1502.03167.

M. T. Ribeiro and C. Guestrin, “‘Why Should I Trust You?’ Explaining the Predictions of Any Classifier Marco,” Proc. 22nd ACM SIGKDD Int. Conf. Knowl. Discov. Data Min., pp. 1135–1144, 2016, doi: 10.1145/2939672.2939778.

W. Hryniewska, A. Grudzień, and P. Biecek, “LIMEcraft: handcrafted superpixel selection and inspection for Visual eXplanations,” Mach. Learn., no. 0123456789, 2022, doi:10.1007/s10994-022-06204-w.

B. K. Law and L. P. Lin, “Development Of A Deep Learning Model To Classify X-Ray Of Covid-19, Normal And Pneumonia-Affected Patients,” Proc. 2021 IEEE Int. Conf. Signal Image Process. Appl. ICSIPA 2021, pp. 1–6, 2021, doi:10.1109/icsipa52582.2021.9576804.

N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov, “Dropout: A simple way to prevent neural networks from overfitting,” J. Mach. Learn. Res., vol. 15, pp. 1929–1958, 2014.

Y. Sari, Y. F. Arifin, Novitasari, and M. R. Faisal, “The Effect of Batch Size and Epoch on Performance of ShuffleNet-CNN Architecture for Vegetation Density Classification,” ACM Int. Conf. Proceeding Ser., pp. 39–46, 2022, doi: 10.1145/3568231.3568239.

D. P. Kingma and J. L. Ba, “Adam: A method for stochastic optimization,” 3rd Int. Conf. Learn. Represent. ICLR 2015, pp. 1–15, 2015, doi: 10.48550/arXiv.1412.6980.

K. S. Wong and L. P. Lin, “A Comparison of Six Convolutional Neural Networks for Weapon Categorization,” Proc. Int. Conf. Electr. Eng. Informatics, pp. 1–6, 2022, doi:10.1109/iceltics56128.2022.9932092.

Y. Y. Chua and L. P. Lin, “Hyperparameter-Tuned CNN for the Classification of Ten Tomato Plant Diseases,” Proc. - 2022 2nd Int. Conf. Electron. Electr. Eng. Intell. Syst. ICE3IS 2022, pp. 1–6, 2022, doi: 10.1109/ICE3IS56585.2022.10010269.

P. Zawodny, W. Malec, K. Gill, K. Skonieczna-Żydecka, and J. Sieńko, “Assessment of the Effectiveness of Treatment of Vascular Lesions within the Facial Skin with a Laser with a Wavelength of 532 nm Based on Photographic Diagnostics with the Use of Polarized Light,” Sensors, vol. 23, p. 1010, 2023, doi: 10.3390/s23021010.

L. E. Davis, S. C. Shalin, and A. J. Tackett, “Current state of melanoma diagnosis and treatment,” Cancer Biol. Ther., vol. 20, no. 11, pp. 1366–1379, 2019, doi: 10.1080/15384047.2019.1640032.

D. Chicco, N. Tötsch, and G. Jurman, “The matthews correlation coefficient (Mcc) is more reliable than balanced accuracy, bookmaker informedness, and markedness in two-class confusion matrix evaluation,” BioData Min., vol. 14, no. 13, pp. 1–22, 2021, doi: 10.1186/s13040-021-00244-z.

J. Liu and Y. Zhao, “Improved generalization performance of convolutional neural networks with LossDA,” Appl. Intell., vol. 53, no. 11, pp. 13852–13866, 2023, doi: 10.1007/s10489-022-04208-6.

C. Wang, J. Sun, W. Xu, and X. Chen, “Depth learning standard deviation loss function,” J. Phys. Conf. Ser., vol. 1176, no. 032050, pp. 1–7, 2019, doi: 10.1088/1742-6596/1176/3/032050.