Getting the most of few data for neonatal pain assessment

ABSTRACT

In this paper we present three different learning models for the automatic assessment of neonatal pain during clinical procedures. Given that few data are publicly available for the training and evaluation of these systems, we develop solutions that try to get the most out of the data that are available. To accomplish this, we choose a convolutional neural network (CNN) architecture as the discriminator with a reduced number of trainable parameters that are used efficiently. Furthermore, we develop two solutions based on the generative adversarial network (GAN) framework in order to improve discriminator power by transforming it into a multitask classifier and by training it with a combination of real and synthetic samples. Experimental results on the publicly available infant Classification Of Pain Expressions (iCOPE) database show superior results compared to previous works in the state of the art.

References

1. Martin Arjovsky, Soumith Chintala, and Léon Bottou. 2017. Wasserstein gan. arXiv preprint arXiv:1701.07875 (2017).
2. Simone Bianco, Remi Cadene, Luigi Celona, and Paolo Napoletano. 2018. Benchmark Analysis of Representative Deep Neural Network Architectures. IEEE Access 6 (2018), 64270--64277.
3. Sheryl Brahnam, Chao-Fa Chuang, Randall S Sexton, and Frank Y Shih. 2007. Machine assessment of neonatal facial expressions of acute pain. Decision Support Systems 43, 4 (2007), 1242--1254.
4. Sheryl Brahnam, Chao-Fa Chuang, Frank Y Shih, and Melinda R Slack. 2006. Machine recognition and representation of neonatal facial displays of acute pain. Artificial intelligence in medicine 36, 3 (2006), 211--222.
5. Sheryl Brahnam, Loris Nanni, and Randall Sexton. 2007. Introduction to neonatal facial pain detection using common and advanced face classification techniques. Springer, Berlin, Heidelberg, 225--253.
6. Sheryl Brahnam, Loris Nanni, and Randall S Sexton. 2008. Neonatal Facial Pain Detection Using NNSOA and LSVM. In Ipcv. 352--357.
7. Luigi Celona and Luca Manoni. 2017. Neonatal facial pain assessment combining hand-crafted and deep features. In ICIAP. Springer, 197--204.
8. Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. 2014. Generative adversarial nets. In NIPS. 2672--2680.
9. Ishaan Gulrajani, Faruk Ahmed, Martin Arjovsky, Vincent Dumoulin, and Aaron C Courville. 2017. Improved training of wasserstein gans. In NIPS. 5767--5777.
10. Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. 2015. Delving deep into rectifiers: Surpassing human-level performance on imagenet classification. In ICCV. 1026--1034.
11. Davis E. King. 2009. Dlib-ml: A Machine Learning Toolkit. Journal of Machine Learning Research 10 (2009), 1755--1758.
12. Yann LeCun, Yoshua Bengio, and Geoffrey Hinton. 2015. Deep learning. Nature 521, 7553 (2015), 436--444.
13. Muhammad Naufal Mansor and Mohd Nazri Rejab. 2013. A computational model of the infant pain impressions with Gaussian and Nearest Mean Classifier. In ICCSCE. IEEE, 249--253.
14. Loris Nanni, Sheryl Brahnam, and Alessandra Lumini. 2010. A local approach based on a Local Binary Patterns variant texture descriptor for classifying pain states. Expert Systems with Applications 37, 12 (2010), 7888--7894.
15. Augustus Odena, Christopher Olah, and Jonathon Shlens. 2017. Conditional image synthesis with auxiliary classifier gans. In ICML, Vol. 70. JMLR. org, 2642--2651.
16. Adam Paszke, Sam Gross, Soumith Chintala, Gregory Chanan, Edward Yang, Zachary DeVito, Zeming Lin, Alban Desmaison, Luca Antiga, and Adam Lerer. 2017. Automatic differentiation in PyTorch. (2017).
17. Alec Radford, Luke Metz, and Soumith Chintala. 2015. Unsupervised representation learning with deep convolutional generative adversarial networks. arXiv preprint arXiv:1511.06434 (2015).
18. Mark Sandler, Andrew Howard, Menglong Zhu, Andrey Zhmoginov, and Liang-Chieh Chen. 2018. MobileNetV2: Inverted Residuals and Linear Bottlenecks. In CVPR. IEEE, 4510--4520.

Index Terms

• Computing methodologies

  • Artificial intelligence

    • Computer vision

  • Machine learning

    • Learning paradigms

      • Supervised learning

        • Supervised learning by classification

    • Machine learning approaches

      • Neural networks