Development and validation of a psoriasis image database for training artificial intelligence in psoriasis diagnosis
Main Article Content
Abstract
The study aims to develop and validate a clinical image database of psoriasis for training artificial intelligence to support the diagnosis of psoriasis at the National Hospital of Dermatology and Venereology. The database consists of 23,752 images from 1,628 psoriasis patients, of which there are 10,939 images of psoriasis vulgaris, 6,847 images of pustular psoriasis, and 5,966 images of erythrodermic psoriasis. The AI model trained on this database achieved a sensitivity of 93.1% and a specificity of 85.9% for diagnosing psoriasis vulgaris; for pustular psoriasis, the sensitivity was 85.9% and the specificity was 93.1%. The overall accuracy of the model is 90.5%. The study has established a clinical psoriasis image database with a sufficient quantity of images and a diversity of psoriasis types. The testing results show that the AI model is capable of diagnosing psoriasis based on clinical images.
Article Details
Keywords
artificial intelligence, database, diagnosis, psoriasis
References
2. Goessinger EV, Gottfrois P, Mueller AM, et al. Image-Based Artificial Intelligence in Psoriasis Assessment: The Beginning of a New Diagnostic Era? Am J Clin Dermatol. 2024;25(6):861-872. doi:10.1007/s40257-024-00883-y
3. Li H, Chen G, Zhang L, et al. A review of psoriasis image analysis based on machine learning. Front Med. 2024;11:1414582. doi:10.3389/fmed.2024.1414582
4. Willemink MJ, Koszek WA, Hardell C, et al. Preparing Medical Imaging Data for Machine Learning. Radiology. 2020;295(1):4-15. doi:10.1148/radiol.2020192224
5. Tschandl P, Rosendahl C, Kittler H. The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions. Sci Data. 2018;5(1):180161. doi:10.1038/sdata.2018.161
6. Bộ Y tế. Hướng dẫn chẩn đoán và điều trị các bệnh da liễu. Xuất bản lần thứ 2. Nhà xuất bản Y học; 2024.
7. Griffiths CE, Barker JN. Pathogenesis and clinical features of psoriasis. The Lancet. 2007;370(9583):263-271. doi:10.1016/S0140-6736(07)61128-3
8. Gooderham MJ, Van Voorhees AS, Lebwohl MG. An update on generalized pustular psoriasis. Expert Rev Clin Immunol. 2019;15(9):907-919. doi:10.1080/1744666X.2019.1648209
9. Zheng M, Jullien D, Eyerich K. The Prevalence and Disease Characteristics of Generalized Pustular Psoriasis. Am J Clin Dermatol. 2022;23(S1):5-12. doi:10.1007/s40257-021-00664-x
10. Rosenbach M, Hsu S, Korman NJ, et al. Treatment of erythrodermic psoriasis: From the medical board of the National Psoriasis Foundation. J Am Acad Dermatol. 2010;62(4):655-662. doi:10.1016/j.jaad.2009.05.048
11. Riley RD, Ensor J, Snell KIE, et al. Calculating the sample size required for developing a clinical prediction model. BMJ. Published online March 18, 2020:m441. doi:10.1136/bmj.m441
12. Zhao S, Xie B, Li Y, et al. Smart identification of psoriasis by images using convolutional neural networks: a case study in China. J Eur Acad Dermatol Venereol. 2020;34(3):518-524. doi:10.1111/jdv.15965
13. Wu H, Yin H, Chen H, et al. A deep learning, image based approach for automated diagnosis for inflammatory skin diseases. Ann Transl Med. 2020;8(9):581-581. doi:10.21037/atm.2020.04.39
114. Huang K, Jiang Z, Li Y, et al. The Classification of Six Common Skin Diseases Based on Xiangya-Derm: Development of a Chinese Database for Artificial Intelligence. J Med Internet Res. 2021;23(9):e26025. doi:10.2196/26025
15. Aijaz SF, Khan SJ, Azim F, et al. Deep Learning Application for Effective Classification of Different Types of Psoriasis. Hasikin K, ed. J Healthc Eng. 2022;2022:1-12. doi:10.1155/2022/7541583