Multiclass Skin Cancer Classification Using Ensemble of Fine-Tuned Deep Learning Models
Source abstract
Abstract
Skin cancer is a widespread disease associated with eight diagnostic classes. The diagnosis of multiple types of skin cancer is a challenging task for dermatologists due to the similarity of skin cancer classes in phenotype. The average accuracy of multiclass skin cancer diagnosis is 62% to 80%. Therefore, the classification of skin cancer using machine learning can be beneficial in the diagnosis and treatment of the patients. Several researchers developed skin cancer classification models for binary class but could not extend the research to multiclass classification with better performance ratios. We have developed deep learning-based ensemble classification models for multiclass skin cancer classification. Experimental results proved that the individual deep learners perform better for skin cancer classification, but still the development of ensemble is a meaningful approach since it enhances the classification accuracy. Results show that the accuracy of individual learners of ResNet, InceptionV3, DenseNet, InceptionResNetV2, and VGG-19 are 72%, 91%, 91.4%, 91.7% and 91.8%, respectively. The accuracy of proposed majority voting and weighted majority voting ensemble models are 98% and 98.6%, respectively. The accuracy of proposed ensemble models is higher than the individual deep learners and the dermatologists’ diagnosis accuracy. The proposed ensemble models are compared with the recently developed skin cancer classification approaches. The results show that the proposed ensemble models outperform recently developed multiclass skin cancer classification models.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Cutaneous Melanoma Detection and Management · AI in cancer detection · Digital Imaging for Blood Diseases
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Computing and Technology, Iqra University, Islamabad 44000, Pakistan
MIS Division, Pakistan Institute of Nuclear Science and Technology (PINSTECH), Islamabad 44000, Pakistan
School of Informatics and Applied Mathematics, Universiti Malaysia Terengganu (UMT), Terengganu 21030, Malaysia
The Norwegian Colour and Visual Computing Laboratory (ColorLab), Department of Computer Science, Norwegian University of Science and Technology (NTNU), 2815 Gjøvik, Norway
Department of Circulation and Medical Imaging, Faculty of Medicine and Health Sciences, Norwegian University of Science and Technology (NTNU), 7491 Trondheim, Norway
References (57)
Show source-provided references
1. Dermoscopic diagnostic performance of Japanese dermatologists for skin tumors differs by patient origin: A deep learning convolutional neural network closes the gap10.1111/1346-8138.15640 ↗
2. A methodological approach to the classification of dermoscopy images10.1016/j.compmedimag.2007.01.003 ↗
3. Quantitative assessment of tumour extraction from dermoscopy images and evaluation of computer-based extraction methods for an automatic melanoma diagnostic system10.1097/01.cmr.0000215041.76553.58 ↗
4. Overview of advanced computer vision systems for skin lesions characterization10.1109/TITB.2009.2017529 ↗
5. Automatic detection of blue-white veil and related structures in dermoscopy images10.1016/j.compmedimag.2008.08.003 ↗
6. Detection of brain abnormality by a novel Lu-Net deep neural CNN model from MR images
7. Pathological brain detection based on AlexNet and transfer learning10.1016/j.jocs.2018.11.008 ↗
8. Abnormal breast identification by nine-layer convolutional neural network with parametric rectified linear unit and rank-based stochastic pooling10.1016/j.jocs.2018.05.005 ↗
9. Multi-grade brain tumor classification using deep CNN with extensive data augmentation10.1016/j.jocs.2018.12.003 ↗
10. Skin cancer segmentation and classification with improved deep convolutional neural network
11. Dermatologist-level classification of skin cancer with deep neural networks10.1038/nature21056 ↗
12. Detection of skin diseases from dermoscopy image using the combination of convolutional neural network and one-versus-all
13. Ratul, M.A.R., Mozaffari, M.H., Lee, W.S., and Parimbelli, E. (2020). Skin lesions classification using deep learning based on dilated convolution. BioRxiv, 860700.10.1101/860700 ↗
14. The skin cancer classification using deep convolutional neural network10.1007/s11042-018-5714-1 ↗
15. Lopez, A.R., Giro-i Nieto, X., Burdick, J., and Marques, O. (2017, January 20–21). Skin lesion classification from dermoscopic images using deep learning techniques. Proceedings of the 2017 13th IASTED International Conference on Biomedical Engineering (BioMed), Innsbruck, Austria.
16. Jinnai, S., Yamazaki, N., Hirano, Y., Sugawara, Y., Ohe, Y., and Hamamoto, R. (2020). The development of a skin cancer classification system for pigmented skin lesions using deep learning. Biomolecules, 10.10.3390/biom10081123 ↗
17. Adegun, A.A., Viriri, S., and Yousaf, M.H. (2021). A Probabilistic-Based Deep Learning Model for Skin Lesion Segmentation. Appl. Sci., 11.10.3390/app11073025 ↗
18. Skin lesion classification with ensembles of deep convolutional neural networks10.1016/j.jbi.2018.08.006 ↗
19. Majtner, T., Bajić, B., Yildirim, S., Hardeberg, J.Y., Lindblad, J., and Sladoje, N. (2018). Ensemble of convolutional neural networks for dermoscopic images classification. arXiv.
20. Nyíri, T., and Kiss, A. (2018, January 12–14). Novel ensembling methods for dermatological image classification. Proceedings of the International Conference on Theory And Practice of Natural Computing, Dublin, Ireland.10.1007/978-3-030-04070-3_34 ↗
21. Bajwa, M.N., Muta, K., Malik, M.I., Siddiqui, S.A., Braun, S.A., Homey, B., Dengel, A., and Ahmed, S. (2020). Computer-aided diagnosis of skin diseases using deep neural networks. Appl. Sci., 10.10.3390/app10072488 ↗
22. He, K., Zhang, X., Ren, S., and Sun, J. (2016, January 27–30). Deep residual learning for image recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA.10.1109/CVPR.2016.90 ↗
23. Hu, J., Shen, L., and Sun, G. (2018, January 18–23). Squeeze-and-excitation networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA.10.1109/CVPR.2018.00745 ↗
24. An effective adaptive customization framework for small manufacturing plants using extreme gradient boosting-XGBoost and random forest ensemble learning algorithms in an Industry 4.0 environment
25. Dynamic fuzzy rule-based source selection in distributed decision fusion systems10.1080/16168658.2018.1509524 ↗
26. A Bi-fold Approach to Detect and Classify COVID-19 X-ray Images and Symptom Auditor10.1007/s42979-021-00701-w ↗
27. Codella, N., Rotemberg, V., Tschandl, P., Celebi, M.E., Dusza, S., Gutman, D., Helba, B., Kalloo, A., Liopyris, K., and Marchetti, M. (2019). Skin lesion analysis toward melanoma detection 2018: A challenge hosted by the international skin imaging collaboration (ISIC). arXiv.
28. Automated melanoma recognition in dermoscopy images via very deep residual networks10.1109/TMI.2016.2642839 ↗
29. Almaraz-Damian, J.A., Ponomaryov, V., Sadovnychiy, S., and Castillejos-Fernandez, H. (2020). Melanoma and nevus skin lesion classification using handcraft and deep learning feature fusion via mutual information measures. Entropy, 22.10.3390/e22040484 ↗
30. Fusing fine-tuned deep features for skin lesion classification10.1016/j.compmedimag.2018.10.007 ↗
31. Hameed, N., Shabut, A.M., and Hossain, M.A. (2018, January 3–5). Multi-class skin diseases classification using deep convolutional neural network and support vector machine. Proceedings of the 2018 12th International Conference on Software, Knowledge, Information Management & Applications (SKIMA), Phnom Penh, Cambodia.10.1109/SKIMA.2018.8631525 ↗
32. Kawahara, J., and Hamarneh, G. (2016). Multi-resolution-tract CNN with hybrid pretrained and skin-lesion trained layers. International Workshop on Machine Learning in Medical Imaging, Springer.10.1007/978-3-319-47157-0_20 ↗
33. Diagnostic accuracy of dermoscopy10.1016/S1470-2045(02)00679-4 ↗
34. Clinical accuracy of the diagnosis of cutaneous malignant melanoma10.1046/j.1365-2133.1998.02075.x ↗
35. Epiluminescence microscopy: A useful tool for the diagnosis of pigmented skin lesions for formally trained dermatologists10.1001/archderm.1995.01690150050011 ↗
36. Dermoscopic diagnosis by a trained clinician vs. a clinician with minimal dermoscopy training vs. computer-aided diagnosis of 341 pigmented skin lesions: A comparative study10.1046/j.1365-2133.2002.04978.x ↗
37. Computer applications in the diagnosis and prognosis of malignant melanoma10.1016/S0733-8635(18)30374-7 ↗
38. ABCD rule and pre-trained CNNs for melanoma diagnosis10.1007/s11042-018-6404-8 ↗
39. Classification of malignant melanoma and benign skin lesions: Implementation of automatic ABCD rule10.1049/iet-ipr.2015.0385 ↗
40. The ABCD rule of dermatoscopy: High prospective value in the diagnosis of doubtful melanocytic skin lesions10.1016/S0190-9622(94)70061-3 ↗
41. Ballerini, L., Fisher, R.B., Aldridge, B., and Rees, J. (2013). A color and texture based hierarchical K-NN approach to the classification of non-melanoma skin lesions. Color Medical Image Analysis, Springer.10.1007/978-94-007-5389-1_4 ↗
42. Digital image analysis for diagnosis of cutaneous melanoma. Development of a highly effective computer algorithm based on analysis of 837 melanocytic lesions10.1111/j.1365-2133.2004.06210.x ↗
43. Yu, Z., Ni, D., Chen, S., Qin, J., Li, S., Wang, T., and Lei, B. (2017, January 18–21). Hybrid dermoscopy image classification framework based on deep convolutional neural network and Fisher vector. Proceedings of the 2017 IEEE 14th International Symposium on Biomedical Imaging (ISBI 2017), Melbourne, VIC, Australia.10.1109/ISBI.2017.7950524 ↗
44. Deep learning10.1038/nature14539 ↗
45. Computer aided diagnostic support system for skin cancer: A review of techniques and algorithms10.1155/2013/323268 ↗
46. Hosny, K.M., Kassem, M.A., and Foaud, M.M. (2018, January 20–22). Skin cancer classification using deep learning and transfer learning. Proceedings of the 2018 9th Cairo International Biomedical Engineering Conference (CIBEC), Cairo, Egypt.10.1109/CIBEC.2018.8641762 ↗
47. Kawahara, J., BenTaieb, A., and Hamarneh, G. (2016, January 13–16). Deep features to classify skin lesions. Proceedings of the 2016 IEEE 13th International Symposium on Biomedical Imaging (ISBI), Prague, Czech Republic.10.1109/ISBI.2016.7493528 ↗
48. An enhanced technique of skin cancer classification using deep convolutional neural network with transfer learning models
49. Shahin, A.H., Kamal, A., and Elattar, M.A. (2018, January 20–22). Deep ensemble learning for skin lesion classification from dermoscopic images. Proceedings of the 2018 9th Cairo International Biomedical Engineering Conference (CIBEC), Cairo, Egypt.10.1109/CIBEC.2018.8641815 ↗
50. An approach for multiclass skin lesion classification based on ensemble learning10.1016/j.imu.2021.100659 ↗
51. Wang, W. (2008, January 1–8). Some fundamental issues in ensemble methods. Proceedings of the 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence), Hong Kong, China.10.1109/IJCNN.2008.4634108 ↗
52. A survey on ensemble learning for data stream classification10.1145/3054925 ↗
53. Szegedy, C., Ioffe, S., Vanhoucke, V., and Alemi, A.A. (2017, January 4–9). Inception-v4, inception-resnet and the impact of residual connections on learning. Proceedings of the Thirty-First AAAI conference on Artificial Intelligence, San Francisco, CA, USA.10.1609/aaai.v31i1.11231 ↗
54. Kuncheva, L.I. (2014). Combining Pattern Classifiers: Methods and Algorithms, John Wiley & Sons.10.1002/9781118914564 ↗
55. Ensemble based systems in decision making10.1109/MCAS.2006.1688199 ↗
56. Chaturvedi, S.S., Gupta, K., and Prasad, P.S. (2020, January 13–15). Skin lesion analyser: An efficient seven-way multi-class skin cancer classification using mobilenet. Proceedings of the International Conference on Advanced Machine Learning Technologies and Applications, Jaipur, India.10.1007/978-981-15-3383-9_15 ↗
57. Ge, Z., Demyanov, S., Chakravorty, R., Bowling, A., and Garnavi, R. (2017, January 11–13). Skin disease recognition using deep saliency features and multimodal learning of dermoscopy and clinical images. Proceedings of the International Conference on Medical Image Computing and Computer-Assisted Intervention, Quebec City, QC, Canada.10.1007/978-3-319-66179-7_29 ↗