Ir al menú de navegación principal Ir al contenido principal Ir al pie de página del sitio

SECCIÓN C: INGENIERÍAS

Vol. 13 Núm. 2 (2021): Volumen 13 Número 2

COVID-19 ResNet: Red neural residual para la clasificación de la COVID-19 con aumento de imágenes con optimización bayesiana

DOI
https://doi.org/10.18272/aci.v13i2.2288
Enviado
mayo 13, 2021
Publicado
2021-11-22

Resumen

La COVID-19 es una enfermedad infecciosa causada por un nuevo coronavirus llamado SARS-CoV-2. El primer caso apareció en diciembre del 2019 y hasta el momento sigue representando un gran desafío a nivel mundial.  La precisa detección del virus en pacientes COVID-19 positivos es un paso crucial para reducir la propagación de esta enfermedad altamente contagiosa.  En este trabajo se implemente una red neuronal residual convolucional (ResNet) para el diagnostico automatizado de la COVID-19. La ResNet implementada puede clasificar la radiografía del tórax de un paciente en COVID-19 positivo, neumonía causada por otro virus o bacteria, y paciente saludable. Además, para aumentar la precisión del modelo y superar la escasez de imágenes médicas en el set de entrenamiento, se aplica una estrategia de aumento de datos personalizada utilizando la optimización bayesiana en tres pasos. La ResNet propuesta alcanza un 94% de precisión, 95% de sensibilidad y 95% en el F1-score en set de preba. Adicionalmente, presentamos las operaciones de aumento de datos que ayudaron a incrementar el rendimiento de la red neuronal y que pueden ser utilizados por otros investigadores en el desarrollo de modelos para la clasificación de imágenes médicas.

viewed = 245 times

Citas

  1. World Health Organization. (2020). Retrieved February 20, 2021, from https://www.who.int/emergencies/diseases/novel-coronavirus-2019/question-and-answers-hub/q-a-detail/coronavirus-disease-covid-19#:~:text=symptoms
  2. Corman, V., Landt, O., Kaiser, M., Molenkamp, R. M., Chu, D., & Drosten, C. (2020). Detection of 2019 novel coronavirus (2019-nCoV) by real-time RT-PCR. Eurosurveillance, 25(3), 2000045.
  3. Center of Disease ,Control and Prevention. (2021). Retrieved February 19, 2021, from https://www.cdc.gov/coronavirus/2019-ncov/testing/diagnostic-testing.html
  4. Center of Disease, Control and Prevention. (2021, February 2). Retrieved February 20, 2021, from https://www.cdc.gov/coronavirus/2019-ncov/testing/serology-overview.html
  5. Dai, W.-c., Zhang, H.-w., Yu, J., Xu, H.-j., Chen, H., Luo, S.P., Zhang, H., Liang, L.H.,Wu, X.L.,Lie, Y., & Lin, F. (2020). CT Imaging and Differential Diagnosis of Covid-19. Canadian Association of Radiologists Journal, 71(2), 195-200.
  6. Wong, H., Lam, H., Fong, A., leung, S., Chin, T., & Lo, C. (2020). Frequency and distribution of chest radiographic findings in patients positive for COVID-19. Radiology, 296(2), E72-E78.
  7. Wan, Y., Shang, J., Graham, R., Baric, R., & Li, F. (2020). Receptor recognition by novel coronavirus from Wuhan: An analysis based on decadelong structural studies of Sars. Journal of Virology, 94(7).
  8. Radiology, A. C. (s.f.). Recomendations for the use of the Chest Radiography and Computed Tomography (CT) for suspected COVID-19 Infection. Retrieved February 21, 2021, from https://www.acr.org/Advocacy-and-Economics/ACR-Position-Statements/Recommendations-for-Chest-Radiography-and-CT-for-Suspected-COVID19
  9. Lujan-Garcia, J. E., Moreno-Ibarra, M. A., Villuendas-Rey, Y., & Yanez-Marquez, C. (2020). Fast COVID-19 and pneumonia classification using chest X-ray images. Mathematics, 8(9), 1423
  10. RSNA. (2020, April 15). Retrieved February 20, 2021, from https://pubs.rsna.org/doi/10.1148/radiol.2020201393?_ga=2.72325459.1392118184.1617117795-543966827.1617117790
  11. Smith, D., Grenier, J., Batter, C., & Spieler, B. (2020). A Characteristic Chest Radiographic Patter in the Setting of the COVID-19. Radiology: Cardiothoracic Imaging, 2(5), e200280.
  12. Khuzani, A. Z., Heidari, M., & Shariati, S. A. (2020, May 8). COVID-Classifier: An automated machine learning model to assist in the diagnosis of COVID-19 infection in chest x-ray images. medRxiv. Retrieved March 6, 2021, from https://github.com/abzargar/COVID-Classifier/tree/master/dataset/original_images_preprocessed
  13. Dai, W. C., Zhang, H. W., Yu, J., Xu, H. J., Chen, H., Luo, S. P., & Lin, F. (2020). CT Imaging and Differential Diagnosis of COVID-19. Canadian Association of Radiologist Journal, 2(71), 195-200.
  14. Parvathy, V. S., Pothiraj, S., & Sampson, J. (2020). Optimal Deep Neural Network model based multimodality fused medical image classification. Physical Communication, 41.
  15. Godasu, R., Zeng, D., & Sutrave, K. (2020). Transfer learning in medical image classification: Challenges and opportunities. Transfer, 5, 28-2020.
  16. Baldeon-Calisto, M., & Lai-Yuen, S. K. (2020). AdaResU-Net: Multiobjective adaptive convolutional neural network for medical image segmentation. Neurocomputing, 392, 325-340.
  17. Baldeon Calisto, M., & Lai-Yuen, S. K. (2020). AdaEn-Net: An ensemble of adaptive 2D-3D Fully Convolutional Networks for medical image segmentation. Neural Networks , 126, 76-94.
  18. Baldeon Calisto, M., & Lai-Yuen, S. (2021). EMONAS-Net: efficient multiobjective neural architecture search framework for 3D medical image segmentation. Artificial Intelligence in Medicine, 102154.
  19. Gu, J., Wang, Z., Kuen, J., Ma, L., Shahroudy, A., Shuai, B., & Chen, T. (2018). Recent avances in convolutional neural networks. Pattern Recognizition, 77, 354-377.
  20. Hussain, Z., Gimenez, F., Yi, D., & Rubin, D. (2017). Differential data augmentation techniques for medical imaging classification tasks. AMIA Annual Symposium Proceedings.
  21. He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. Proceedings of the IEEE conference on computer vision and pattern recognition.
  22. Aggarwal, P., Vig, R., Bhadhoria, S., & Dethe, C. (2011). Role of segmentation in medical imaging. International Journal of Computer Applications, 975(8887), 54-61.
  23. Müller, D., & Kramer, F. (2021). MIScnn: a framework for medical image segmentation with convolutional neural networks and deep learing. BMC medical imaging, 1-2.
  24. Wang, S., Kang, B., Ma, J., Zeng, X., Xiao, M., Guo, J., Cai, M., Yang, J., Yaodong, L,. Xu, B., & Meng, X. (2021). A deep learning algorithm using CT images to screen for Corona Virus Disease (COVID-19). European Radiology, 1-9.
  25. Sethy, P. K., Behera, S. K., Ratha, P. K., & Biswas, P. (2020). Detection of coronavirus disease (covid-19) Based on Deep Features. PrePrints.
  26. He, X., Shihao, W., Guohao, Y., Jiyong, Z., & Chu, X. (2020). Efficient Multi-objective Evolutionary 3D Neural Architecture Search for COVID-19 Detection with Chest CT Scans. arXiv preprint arXiv:2101.10667, 1-2.
  27. Narin, A., Kaya, C., & Pamuk, Z. (2020). Automatic Detection of Coronavirus Disease (COVID-19) Using X-ray Images and Deep Convolutional Neural Networks. arXiv preprint arXiv:2003.10849, 2-3.
  28. Zeshan, H., Gimenez, F., Yi, D., & Rubin, D. (2018). Differential Data Augmentation Techniques for Medical Imaging Classification Tasks. Annual Symposium. San Francisco.
  29. Bergstra, J., Barnedet, R., Bengio, Y., & Kegl, B. (2011). Algorithms for hyper-parameter optimization. In Advances in Neural Information Processing Systems (NIPS), 2546-2554.
  30. Bengio, Y. (2012). Practical Recommendations for Gradient-Based Training of Deep Architectures. Neural Networks: Tricks of the trade, 437-478.
  31. Masters, D., & Carlo, L. (2018). Revisiting small batch training for deep neural networks. arXiv preprint arXiv:1804.07612}, 2-3.
  32. Snoek, J., Rippel, O., Swersky, K., Kiros, R., Satish, N., Sundaram, N., Mostofa, P., Prabhat, Mr., & Adams, R. (2015). Scalable bayesian optimization using deep neural networks. International conference on machine learning.
  33. The GPyOpt authors. (2016). GPyOpt: A Bayesian Optimization framework in Python. http://github.com/SheffieldML/GPyOpt
  34. Kaur, S., & Kaur, S. (2014). An EfficientApproach for Number Plate Extraction from Vehicles Image under Image Processing Article. Jalandhar: CT Group of Institutions.
  35. Macedo, S., Givanio, M., & Keiner, J. (2015). A comparative study of grayscale conversion techniques applied to SIFT descriptors. Brazil: Universidade Federal de Pernambuco.
  36. Sherrier, R. H., & Jonhson , G. A. (1987). Regionally Adaptative Histogram Equalization of the Chest. IIEE Transaction on Medical Imaging, 6(1), 1-7.
  37. Kingma, D. P., & Ba, J. (2014). Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980.
  38. Gu, S., Pednekar, M., & Slater, R. (2019). Improve Image Classification Using Data Augmentation and Neural Networks. SMU Data Science Review, 2, 1.
  39. Maghdid, H., Assad, K., Ghafoor, A., & Khan, M. (2020). Diagnosing covid-19 pneumonia from x-ray and ct images using deep learning and transfer learning algorithms. arXiv, 00038(2004).
  40. Farooq, M., & Hafeez, A. (2020). Covid-resnet: A deep learning framework for screening of covid19 from radiographs. arXiv, 2003(14395).
  41. Wang, L., Lin, Z. Q., & Wong, A. (2020). Covid-net: A tailored deep convolutional neural network design for detection of covid-19 cases from chest x-ray images. Scientific Reports, 10(09871), 1-12.
  42. Wu, X., Hui, H., Niu, M., Li, L., Wang, B., Li, H., & Tian, J. (2020). Deep learning-based multi-view fusion model for screening 2019 novel coronavirus pneumonia: a multicentre study,. European Journal of Radiology, 128, 109041.

Descargas

Los datos de descargas todavía no están disponibles.