Support for Skin Cancer Diagnosis Using Artificial Intelligence: Pilot Study
DOI:
https://doi.org/10.32635/2176-9745.RBC.2026v72n1.5443Keywords:
Skin Neoplasms/classification, Degloving Injuries/ classification, Deep Learning, Convolutional Neural Networks, Image Processing, Computer-AssistedAbstract
Introduction: Skin cancer is one of the most prevalent neoplasms in Brazil, and early diagnosis is a key determinant of therapeutic success and reduction of associated morbidity and mortality. In this epidemiological context, there is a particularly favorable scenario for the incorporation of computational tools to complement traditional clinical evaluation, with emphasis on artificial intelligence–based approaches. Objective: To develop and validate a convolutional neural network–based model for the automatic classification of malignant and benign skin lesions. Method: A total of 2,639 images from the public International Skin Imaging Collaboration (ISIC) database, with biopsy-validated annotations, were used. The computational system included preprocessing steps and supervised training using the YOLOv11 architecture. Performance was assessed through internal and external validation. Results: The model achieved a mean accuracy of 80.53% and a mean sensitivity of 80.44% in the identification of eight classes of lesions: melanoma, nevus, basal cell carcinoma, actinic keratosis, benign keratosis, dermatofibroma, vascular lesion, and squamous cell carcinoma. The implementation also resulted in an annotated image dataset and a reproducible analysis pipeline. Conclusion: The application of artificial intelligence to support skin cancer diagnosis demonstrated promising performance, with potential clinical screening applications. Future studies should consider expanding the dataset and developing user interfaces for healthcare professionals.
Downloads
References
Delgado LP, Regis MN, Caldas D, et al. Epidemiologia e manejo clínico do câncer de pele melanoma: uma revisão narrativa. J Soc Issues Health Sci. 2024;1(7):1-11. doi: https://doi.org/10.5281/zenodo.17473590
Rezende Filho AV, Yamamoto HG, Macedo JLS et al. Perfil epidemiológico de pacientes portadores de câncer de pele atendidos no Hospital Regional da Asa Norte/ DF - Brasil. Rev Bras Cir Plást. 2020;35(3):316-21. doi: https://doi.org/10.5935/2177-1235.2020RBCP0056 DOI: https://doi.org/10.5935/2177-1235.2020RBCP0056
Borges AL, Zalaudeck I, Longo C, et al. Melanocytic nevi with special features: clinical-dermoscopic and reflectance confocal microscopic-findings. JEADV. 2014;28(7):833-45. doi: https://doi.org/10.1111/ jdv.12291 DOI: https://doi.org/10.1111/jdv.12291
Piccolo V, Russo T, Moscarella E, et al. Dermatoscopy of vascular lesions. Dermatol Clin. 2018;36(4):389-95. doi: https://doi.org/10.1016/j.det.2018.05.006 DOI: https://doi.org/10.1016/j.det.2018.05.006
Álvarez-Salafranca M, Gómez-Martíns I, Bañuls J, et al. Dermoscopy of inflamed seborrheic keratosis: a great mimic of malignancy. Australas J Dermatol. 2022;63(1):53-61. doi: https://doi.org/10.1111/ajd.13781 DOI: https://doi.org/10.1111/ajd.13781
Thamm JR, Welzel J, Schuh S. Diagnosis and therapy of actinic keratosis. J Dtsch Dermatol Ges. 2024;22(5):675-90. doi: https://doi.org/10.1111/ddg.15288 DOI: https://doi.org/10.1111/ddg.15288
Azulay RA, Azulay DR, Azulay-ABULAFIA L. Dermatologia. 7.ed. Rio de Janeiro: Guanabara Koogan; 2017.
Botton DV, Barbosa DGR, Cavalcante Junior CAC, et al. Relevância da dermatoscopia para o diagnóstico precoce de melanomas: uma revisão de literatura. RISE. 2020;1(2):159-74. doi: https://doi.org/10.56344/2675-4827.v1n2a20209 DOI: https://doi.org/10.56344/2675-4827.v1n2a20209
Schmitt JV, Miot HA. Distribution of Brazilian dermatologists according to geographic location, population and HDI of municipalities: an ecological study. An Bras Dermatol. 2014;89(6):1013-5. doi: https://doi.org/10.1590/abd1806-4841.20143276 DOI: https://doi.org/10.1590/abd1806-4841.20143276
Siqueira ASE, Santos Neto MF, Ferreira CBT, et al. Inteligência artificial nas ações de controle do câncer: solução ou problema? Rev Bras Cancerol. 2025;71(3):e- 005291. doi: https://doi.org/10.32635/2176-9745.RBC.2025v71n3.5291 DOI: https://doi.org/10.32635/2176-9745.RBC.2025v71n3.5291
Phillips M, Marsden H, Jaffe W, et al. Assessment of accuracy of an artificial intelligence algorithm to detect melanoma in images of skin lesions. JAMA Netw Open. 2019;2(10):e1913436. doi: https://doi.org/10.1001/jamanetworkopen.2019.13436 DOI: https://doi.org/10.1001/jamanetworkopen.2019.13436
Munjal G, Bhardwaj P, Bhargava V, et al. SkinSage XAI: an explainable deep learning solution for skin lesion diagnosis. Health Care Sci. 2024;3(6):438-55. doi: https://doi.org/10.1002/hcs2.121 DOI: https://doi.org/10.1002/hcs2.121
Cui X, Wei R, Gong L, et al. Assessing the effectiveness of artificial intelligence methods for melanoma: a retrospective review. J Am Acad Dermatol. 2019;81(5):1176-80. doi: https://doi.org/10.1016/j.jaad.2019.06.042 DOI: https://doi.org/10.1016/j.jaad.2019.06.042
Jairath N, Pahalyants V, Shah R, et al. Artificial intelligence in dermatology: a systematic review of its applications in melanoma and keratinocyte carcinoma diagnosis. Dermatol Surg. 2024;50(9):791-8. doi: https://doi.org/10.1097/DSS.0000000000004223 DOI: https://doi.org/10.1097/DSS.0000000000004223
Wang HH, Wang YH, Liang CW, et al. Assessment of deep learning using nonimaging information and sequential medical records to develop a prediction model for nonmelanoma skin cancer. JAMA Dermatol. 2019;155(11):1277-83. doi: https://doi.org/10.1001/jamadermatol.2019.2335 DOI: https://doi.org/10.1001/jamadermatol.2019.2335
Chu YS, An HG, Oh BH, et al. Artificial intelligence in cutaneous oncology. Front Med (Lausanne). 2020;7:318. doi: https://doi.org/10.3389/fmed.2020.00318 DOI: https://doi.org/10.3389/fmed.2020.00318
Silva TAM, Souza EP, Bambil D, et al. Aprendizado de máquina aplicado ao diagnóstico de câncer de pele. In: 16 Anais do Congresso Brasileiro de Inteligência Computacional [Internet]; 2021 out 3-6; Joinville. Joinville: UDESC; 2021. doi: http://dx.doi.org/10.21528/CBIC2021-55 DOI: https://doi.org/10.21528/CBIC2021-55
Heinlein L, Maron RC, Hekler A, et al. Clinical melanoma diagnosis with artificial intelligence: insights from a prospective multicenter study [Preprint]. arXiv. 2024. doi: https://doi.org/10.48550/arXiv.2401.14193
ISIC 2019: Skin lesion analysis towards melanoma detection [Internet]. Versão 2019. [sem local]: International Skin Imaging Collaboration; 2019 [acesso 2024 set 18]. Disponível em: https://challenge2019.isic-archive.com/
LabelImg [Internet]. Versão 1.8.6. [sem loca]: Python Software Foundation; ©2025 [acesso 2024 set 18]. Disponível em: https://pypi.org/project/labelImg/
Ultralytics. Ultralytics YOLO [repositório GitHub] [acesso 2025 abr 10]. Disponível em: https://github.com/ultralytics/ultralytics
Terven J, Córdova-Esparza DM, Romero-González JA. A comprehensive review of YOLO architectures in computer vision: from YOLOv1 to YOLOv8 and YOLO-NAS. Mach Learn Knowl Extr. 2023;5(4):1680- 716. doi: https://doi.org/10.3390/make5040083 DOI: https://doi.org/10.3390/make5040083
Esteva A, Chou K, Yeung S, et al. Deep learning-enabled medical computer vision. npj Digit Med. 2021;4:5. doi: https://doi.org/10.1038/s41746-020-00376-2 DOI: https://doi.org/10.1038/s41746-020-00376-2
Cechinel C, Camargo SS. Mineração de dados educacionais: avaliação e interpretação de modelos de classificação [Internet]. In: Jaques P, Pimentel M, Siqueira S, et al, editores. Metodologia de pesquisa científica em informática na educação: abordagem quantitativa. Porto Alegre: SBC; 2020. Disponível em: https://ceie.sbc.org.br/metodologia/wp-content/uploads/2019/11/livro2_cap12.pdf
Steyerberg EW, Harrell Jr FE. Prediction models need appropriate internal, internal-external, and external validation. J Clin Epidemiol. 2016;69:245-7. doi: https://doi.org/10.1016/j.jclinepi.2015.04.005 DOI: https://doi.org/10.1016/j.jclinepi.2015.04.005
Saturn Cloud [Internet]. Nova Iorque: Saturn Cloud, ©2025 [acesso 2024 set 18]. Disponível em: https://saturncloud.io/
Conselho Nacional de Saúde (BR). Resolução n° 466, de 12 de dezembro de 2012. Aprova as diretrizes e normas regulamentadoras de pesquisas envolvendo seres humanos [Internet]. Diário Oficial da União, Brasília, DF. 2013 jun 13 [acesso 2025 ago 27]; Seção 1:59. Disponível em: https://bvsms.saude.gov.br/bvs/saudelegis/cns/2013/res0466_12_12_2012.html
Conselho Nacional de Saúde (BR). Resolução n° 510, de 7 de abril de 2016. Dispõe sobre as normas aplicáveis a pesquisas em Ciências Humanas e Sociais cujos procedimentos metodológicos envolvam a utilização de dados diretamente obtidos com os participantes ou de informações identificáveis ou que possam acarretar riscos maiores do que os existentes na vida cotidiana, na forma definida nesta Resolução [Internet]. Diário Oficial da União, Brasília, DF. 2016 maio 24 [acesso 2024 abr 7]; Seção 1:44. Disponível em: http://bvsms.saude.gov.br/bvs/saudelegis/cns/2016/res0510_07_04_2016.html
Published
How to Cite
Issue
Section
License
Os direitos morais e intelectuais dos artigos pertencem aos respectivos autores, que concedem à RBC o direito de publicação.
This work is licensed under a Creative Commons Attribution 4.0 International License.
