Vol 6, No 2 (2020)
Research paper
Published online: 2020-07-09

open access

Page views 666
Article views/downloads 50
Get Citation

Connect on Social Media

Connect on Social Media

Can convolutional neural networks outperform clinicians in the detection of melanoma on dermoscopy images?

Dominika Kwiatkowska1, Piotr Kluska2, Adam Reich1
Forum Dermatologicum 2020;6(2):40-42.

Abstract

Introduction: Artificial intelligence is widely used in various _elds of medicine. It also has great potential for being used in the assessment of dermoscopy images. This study aimed to evaluate whether a convolutional neural network model could match dermatologists’ accuracy in the assessment of dermoscopic pictures.  Material and methods: For this research we used HAM10000 training dataset, that was extracted from “ISIC 2018: Skin Lesion Analysis Towards Melanoma Detection”. All skin lesions were classi_ed to one of the following group: (1) malignant melanoma, (2) melanocytic nevus, (3) basal cell carcinoma, (4) actinic keratosis/Bowen’s disease, (5) benign keratosis, (6) dermato_broma, and (7) vascular lesion. From the dataset, we have randomly extracted 104 images from all classes of lesions to create the online test presented to 14 dermatologists who were asked to classify each lesion out of 104 dermoscopic pictures to the groups mentioned above. Next, the ResNeXt model was evaluated on the same dataset.  Results: Dermatologists achieved better sensitivity than ResNeXt in malignant melanoma di_erentiation. However, precision and F1 score of ResNeXt were higher in comparison to dermatologists. Moreover, CNN was more precise and sensitive to other skin lesion types.  Conclusions: This research has shown that computer vision aided dermoscopy can be a supportive tool that could help physicians in the screening of patients for malignant melanoma. 

Article available in PDF format

View PDF Download PDF file

References

  1. Bacchi S, Zerner T, Dongas J, et al. Deep learning in the detection of high-grade glioma recurrence using multiple MRI sequences: A pilot study. Journal of Clinical Neuroscience. 2019; 70: 11–13.
  2. Badgeley M, Zech J, Oakden-Rayner L, et al. Deep learning predicts hip fracture using confounding patient and healthcare variables. npj Digital Medicine. 2019; 2(1).
  3. Arcadu F, Benmansour F, Maunz A, et al. Deep learning algorithm predicts diabetic retinopathy progression in individual patients. npj Digital Medicine. 2019; 2(1).
  4. Lee C, Baughman D, Lee A. Deep Learning Is Effective for Classifying Normal versus Age-Related Macular Degeneration OCT Images. Ophthalmology Retina. 2017; 1(4): 322–327.
  5. Komura D, Ishikawa S. Machine Learning Methods for Histopathological Image Analysis. Computational and Structural Biotechnology Journal. 2018; 16: 34–42.
  6. LeCun Y, Bengio Y, Hinton G. Deep learning. Nature. 2015; 521(7553): 436–444.
  7. Haenssle HA, Fink C, Schneiderbauer R, et al. Man against machine: diagnostic performance of a deep learning convolutional neural network for dermoscopic melanoma recognition in comparison to 58 dermatologists. Annals of Oncology. 2018; 29(8): 1836–1842.
  8. Sagar A, Dheeba J. Convolutional Neural Networks for Classifying Melanoma Images. .
  9. Tschandl P, Rosendahl C, Kittler H. The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions. Scientific Data. 2018; 5(1).
  10. Deng J, Dong W, Socher R, et al. ImageNet: A large-scale hierarchical image database. 2009 IEEE Conference on Computer Vision and Pattern Recognition. 2009.
  11. Xie S, Girshick R, Dollar P, et al. Aggregated Residual Transformations for Deep Neural Networks. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2017.
  12. Kingma DP, Ba J. Adam: A method for stochastic optimization. arXiv Prepr arXiv1412. 6980; 2014.
  13. Tschandl P, Codella N, Akay B, et al. Comparison of the accuracy of human readers versus machine-learning algorithms for pigmented skin lesion classification: an open, web-based, international, diagnostic study. The Lancet Oncology. 2019; 20(7): 938–947.
  14. Gaudy-Marqueste C, Wazaefi Y, Bruneu Y, et al. Ugly Duckling Sign as a Major Factor of Efficiency in Melanoma Detection. JAMA Dermatology. 2017; 153(4): 279.