The Role of Convolutional Neural Network (CNN) Based on Dermoscopy Imaging for Early Detection of Melanoma: A Systematic Review
Main Article Content
Abstract
Background: AI systems for melanoma detection have shown considerable potential. Although, dermoscopy has become a widely utilized non-invasive method for diagnosing skin tumors. However, the variability in diagnosis caused by subjective interpretation of dermatological findings can affect both accuracy and consistency. Therefore, we conducted this study to review the accuracy, sensitivity, and specificity of Convolutional Neural Network (CNN) based on dermoscopy imaging in diagnosing melanoma.
Method: This systematic review was conducted in accordance with the Preferred Reporting Items of Systematic Reviews (PRISMA) guidelines. We limited the studies from 2019 until 2024. All studies that assessed diagnostic accuracy of CNN in diagnosing melanoma were analyzed. QUADAS (Quality Assessment of Diagnostic Accuracy Studies) is used to assess the quality of diagnostic accuracy studies.
Results: Eleven studies were eligible to be included in this study. The Area Under Curve (AUC) among the studies varied between 81.3% and 92.6%. Sensitivity varied between 69.1% and 94.2%. Specificity varied between 65% and 84.63%.
Conclusion: The AUC, sensitivity, and specificity showed good results compared to dermoscopy alone. However, the usage of artificial intelligence was as an adjunctive tool, not as a replacement for dermatologists.
Article Details
This work is licensed under a Creative Commons Attribution 4.0 International License.
References
I. Saginala K, Barsouk A, Aluru JS, et al. Epidemiology of Melanoma. Medical Sciences 2021;9(4):63;
II. Kato J, Horimoto K, Sato S, et al. Dermoscopy of Melanoma and Non-melanoma Skin Cancers. Front Med (Lausanne) 2019;6
III. Reddy S, Shaheed A, Patel R. Artificial Intelligence in Dermoscopy: Enhancing Diagnosis to Distinguish Benign and Malignant Skin Lesions. Cureus 2024
IV. Heinlein L, Maron RC, Hekler A, et al. Prospective multicenter study using artificial intelligence to improve dermoscopic melanoma diagnosis in patient care. Communications Medicine 2024;4(1):177
V. Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021;372:n71
VI. Foahom Gouabou AC, Collenne J, Monnier J, et al. Computer Aided Diagnosis of Melanoma Using Deep Neural Networks and Game Theory: Application on Dermoscopic Images of Skin Lesions. Int J Mol Sci 2022;23(22):13838
VII. Martin-Gonzalez M, Azcarraga C, Martin-Gil A, et al. Efficacy of a Deep Learning Convolutional Neural Network System for Melanoma Diagnosis in a Hospital Population. Int J Environ Res Public Health 2022;19(7)
VIII. Kaur R, GholamHosseini H, Sinha R, et al. Melanoma Classification Using a Novel Deep Convolutional Neural Network with Dermoscopic Images. Sensors 2022;22(3):1134
IX. Xing X, Song P, Zhang K, et al. ZooME: Efficient Melanoma Detection Using Zoom-in Attention and Metadata Embedding Deep Neural Network. In: 2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC) IEEE; 2021; pp. 4041–4044
X. Kim C-I, Hwang S-M, Park E-B, et al. Computer-Aided Diagnosis Algorithm for Classification of Malignant Melanoma Using Deep Neural Networks. Sensors 2021;21(16):5551
XI. Foahom Gouabou AC, Damoiseaux J-L, Monnier J, et al. Ensemble Method of Convolutional Neural Networks with Directed Acyclic Graph Using Dermoscopic Images: Melanoma Detection Application. Sensors 2021;21(12):3999
XII. R. K, H. G, R. S. Deep Convolutional Neural Network for Melanoma Detection Using Dermoscopy Images. In: 2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC) IEEE; 2020; pp. 1524–1527
XIII. Winkler JK, Sies K, Fink C, et al. Melanoma recognition by a deep learning convolutional neural network—Performance in different melanoma subtypes and localisations. Eur J Cancer 2020;127:21–29
XIV. Brinker TJ, Hekler A, Enk AH, et al. Deep neural networks are superior to dermatologists in melanoma image classification. Eur J Cancer 2019;119:11–17
XV. Maron RC, Weichenthal M, Utikal JS, et al. Systematic outperformance of 112 dermatologists in multiclass skin cancer image classification by convolutional neural networks. Eur J Cancer 2019;119:57–65
XVI. Papageorgiou V, Apalla Z, Sotiriou E, et al. The limitations of dermoscopy: false‐positive and false‐negative tumours. Journal of the European Academy of Dermatology and Venereology 2018;32(6):879–888; doi
XVII. Bian X, Pan H, Zhang K, et al. Malignant melanoma dermoscopy image classification method based on multi‐modal medical features. IET Image Process 2023;17(9):2611–2627
XVIII. Zhang S, Wang Y, Zheng Q, et al. Artificial intelligence in melanoma: A systematic review. J Cosmet Dermatol 2022;21(11):5993–6004
XIX. Zhao X, Wang L, Zhang Y, et al. A review of convolutional neural networks in computer vision. Artif Intell Rev 2024;57(4):99
XX. Yamashita R, Nishio M, Do RKG, et al. Convolutional neural networks: an overview and application in radiology. Insights Imaging 2018;9(4):611–629
XXI. Patel RH, Foltz EA, Witkowski A, et al. Analysis of Artificial Intelligence-Based Approaches Applied to Non-Invasive Imaging for Early Detection of Melanoma: A Systematic Review. Cancers (Basel) 2023;15(19):4694
XXII. Marchetti MA, Cowen EA, Kurtansky NR, et al. Prospective validation of dermoscopy-based open-source artificial intelligence for melanoma diagnosis (PROVE-AI study). NPJ Digit Med 2023;6(1):127
XXIII. Holmes GA, Vassantachart JM, Limone BA, et al. Using Dermoscopy to Identify Melanoma and Improve Diagnostic Discrimination. Fed Pract 2018;35(Suppl 4):S39–S45.