Evaluación de variaciones de la red EfficientNet en conjuntos dermatoscópicos

Autores/as

  • Newton Spolaôr Universidade Estadual do Oeste do Paraná
  • Huei Diana Lee UNIOESTE
  • Weber Shoity Resende Takaki Universidade Estadual do Oeste do Paraná
  • Claudio Saddy Rodrigues Coy UNICAMP
  • Feng Chung Wu UNIOESTE

DOI:

https://doi.org/10.59681/2175-4411.v16.iEspecial.2024.1337

Palabras clave:

Inteligencia Artificial, Informática Médica, Neoplasias Cutáneas

Resumen

Objetivo: Investigar configuraciones pioneras de la red profunda EfficientNet-B2 para clasificar pequeñas bases de datos dermatoscópicas. Método: Un enfoque para (1) preprocesamiento de imágenes, (2) clasificación con ocho configuraciones para ajustar una EfficientNet-B2 previamente entrenada y (3) evaluación de clasificadores con validación cruzada estratificada en tres bases de datos dermatoscópicas. Resultados: Todos los modelos superaron la línea de base. Se encontraron algunas diferencias estadísticas entre ellos. La mejor red alcanzó una precisión promedia de 98,33% en el conjunto público PH2. Conclusión: Algunas configuraciones pioneras de la red profunda fueron competitivas frente a referencias recientes en la clasificación de dermatoscopias.

Biografía del autor/a

Newton Spolaôr, Universidade Estadual do Oeste do Paraná

Doutor, Laboratório de Bioinformática, Universidade Estadual do Oeste do Paraná – LABI/UNIOESTE, Foz do Iguaçu (PR), Brasil.

Huei Diana Lee, UNIOESTE

Professor Associado-III Doutor, LABI/UNIOESTE, Foz do Iguaçu (PR), Brasil.

Weber Shoity Resende Takaki, Universidade Estadual do Oeste do Paraná

Doutor, Laboratório de Bioinformática, Universidade Estadual do Oeste do Paraná – LABI/UNIOESTE, Foz do Iguaçu (PR), Brasil.

Claudio Saddy Rodrigues Coy, UNICAMP

Professor Titular Doutor, Faculdade de Ciências Médicas, Universidade Estadual de Campinas – FCM/UNICAMP, Campinas (SP), Brasil.

Feng Chung Wu, UNIOESTE

Professor Associado-III Doutor, LABI/UNIOESTE, Foz do Iguaçu (PR), Brasil.

Citas

Malik FS, Yousaf MH, Sial HA, Viriri S. Exploring dermoscopic structures for melanoma lesions’ classification. Front Big Data. 2024;7:1366312.

Spolaôr N, Lee HD, Mendes AI, Nogueira CV, Parmezan ARS, Takaki WSR, et al. Fine-tuning pre-trained neural networks for medical image classification in small clinical datasets. Multimed Tools Appl. 2024;83(9):27305-29.

Balaha HM, Hassan AES. Skin cancer diagnosis based on deep transfer learning and sparrow search algorithm. Neural Comput Appl. 2023;35(1):815-53.

Instituto Nacional de Câncer (BR). Estimativa 2023: incidência de câncer no Brasil [Internet]. Rio de Janeiro: Instituto Nacional de Câncer; 2023 [citado 2024 Mai 22]. Disponível em: https://www.inca.gov.br/publicacoes/livros/estimativa-2023-incidencia-de-cancer-no-brasil.

Venugopal V, Raj NI, Nath MK, Stephen N. A deep neural network using modified EfficientNet for skin cancer detection in dermoscopic images. Decision Analytics Journal. 2023;8:100278.

Bansal P, Garg R, Soni P. Detection of melanoma in dermoscopic images by integrating features extracted using handcrafted and deep learning models. Comput Ind Eng. 2022;168:108060.

Hasan Rafi T, Shubair RM. A scaled-2D CNN for skin cancer diagnosis. In: Hallinan J, Chetty M, Heredia GR, et al., editors. Proceedings of the 18th IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology; 2021; Melbourne, Australia. [New York]: Curran Associates; 2021. p. 1-6.

Chollet F, Kalinowski T, Allaire JJ. Deep learning in R. 2nd ed. Shelter Island: Manning publications; 2022.

Liu XJ, Li Kl, Luan Hy, Wang Wh, Chen Zy. Few-shot learning for skin lesion image classification. Multimed Tools Appl. 2022;81(4):4979-90.

Tan M, Le QV. EfficientNet: Rethinking model scaling for convolutional neural networks. In: Chaudhuri K, Salakhutdinov R, editors. Proceedings of the 36th International Conference on Machine Learning; 2019; Long Beach, United States. [Brookline]: [Microtome Publishing]; 2019. p. 6105-14.

Jaisakthi SM, Mirunalini P, Aravindan C, Appavu R. Classification of skin cancer from dermoscopic images using deep neural network architectures. Multimed Tools Appl. 2023;82(10):15763-78.

Tajerian A, Kazemian M, Tajerian M, Akhavan Malayeri A. Design and validation of a new machine-learning-based diagnostic tool for the differentiation of dermatoscopic skin cancer images. PLoS One. 2023;18(4):1-17.

Papiththira S, Kokul T. Melanoma skin cancer detection using EfficientNet and channel attention module. In: Wijayakulasooriya J, editor. Proceedings of the 16th IEEE International Conference on Industrial and Information Systems; 2021; Kandy, Sri Lanka. [New York]: Curran Associates; 2021. p. 227-32.

Abadi M, Agarwal A, Barham P, Brevdo E, Chen Z, Citro C, et al. TensorFlow: Large-scale machine learning on heterogeneous systems. Version 2.14 [software]. 2023 [cited 2024 May 22]. Available from: http://tensorflow.org.

Lee HD, Mendes AI, Spolaôr N, Oliva JT, Sabino Parmezan AR, Chung WF, et al. Dermoscopic assisted diagnosis in melanoma: Reviewing results, optimizing methodologies and quantifying empirical guidelines. Knowl Based Syst. 2018;158:9-24.

Machado M, Pereira J, Fonseca-Pinto R. Classification of reticular pattern and streaks in dermoscopic images based on texture analysis. J Med Imaging. 2015;2(4):044503.

Argenziano G, Zalaudek I. Dermoscopy: a new perspective. Dermatol Pract Concept. 2011;1(1):57-8.

Boer A, Nischal K. A growing online resource for learning dermatology and dermatopathology. Indian J Dermatol Venereol Leprol. 2007;73(2):138-40.

Mendonça TF, Ferreira PM, Marçal ARS, Barata C, Marques JS, Rocha J, et al. PH2: A public database for the analysis of dermoscopic images. In: Celebi ME, Mendonça TF, Marques JS, editors. Dermoscopy Image Analysis. Boca Ratón: CRC Press; 2016. p. 419-40.

Codella N, Rotemberg V, Tschandl P, Celebi ME, Dusza S, Gutman D, et al. Skin lesion analysis toward melanoma detection 2018: A challenge hosted by the international skin imaging collaboration (ISIC). arXiv: 1902.03368 [Preprint]. 2019 [cited 2024 May 22]: [12 p.]. Available from: https://arxiv.org/abs/1902.03368.

Sandler M, Howard A, Zhu M, Zhmoginov A, Chen LC. MobileNetV2: Inverted residuals and linear bottlenecks. In: Brown MS, Morse B, Peleg S, editors. Proceedings of the 31st IEEE Conference on Computer Vision and Pattern Recognition; 2018; Salt Lake City, United States. [Washington]: IEEE Computer Society; 2018. p. 4510-20.

Witten IH, Frank E, Hall MA, Pal CJ. Data mining: Practical machine learning tools and techniques. 4th ed. Burlington: Morgan Kaufmann; 2016.

Velden BHM, Kuijf HJ, Gilhuijs KGA, Viergever MA. Explainable artificial intelligence (XAI) in deep learning-based medical image analysis. Med Image Anal. 2022;79:102470.

Chougrad H, Zouaki H, Alheyane O. Deep convolutional neural networks for breast cancer screening. Comput Methods Programs Biomed. 2018;157:19-30.

Publicado

2024-11-19

Cómo citar

Spolaôr, N., Lee, H. D., Takaki, W. S. R., Coy, C. S. R., & Wu, F. C. (2024). Evaluación de variaciones de la red EfficientNet en conjuntos dermatoscópicos. Journal of Health Informatics, 16(Especial). https://doi.org/10.59681/2175-4411.v16.iEspecial.2024.1337

Artículos similares

<< < 4 5 6 7 8 9 10 11 12 13 > >> 

También puede {advancedSearchLink} para este artículo.

Artículos más leídos del mismo autor/a