Automated pain recognition from facial movements of newborns admitted to Neonatal Intensive Care Units

• Tatiany Marcondes Heiderich FEI
• Carlos Eduardo Thomaz FEI

DOI: https://doi.org/10.5753/sbcas_estendido.2025.6903

Abstract

This thesis aimed to develop an automated method for recognizing pain in neonates using state-of-the-art image processing and artificial intelligence techniques. The method segments facial regions and classifies pain-indicative movements. Analyzed in four datasets, the method showed promising results, with the nasolabial furrow deepening movement standing out as the most reliable region for pain diagnosis. A new scoring method for automatic pain assessment was further proposed and demonstrated high reliability. This new approach optimizes traditional methods, providing a more objective and precise neonatal pain assessment in complex clinical scenarios.

References

Batton, D., Barrington, K. and Wallman, C. (2006) “Prevention and Management of Pain in the Neonate: An Update” Pediatrics, v. 118, n. 5, p. 2231–2241.

Brahnam, S., Nanni, L., McMurtrey, S., et al. (2020) “Neonatal pain detection in videos using the iCOPEvid dataset and an ensemble of descriptors extracted from Gaussian of Local Descriptors” Applied Computing and Informatics, p. 1–22.

Buzuti, L., Heideirich, T., Barros, M., Guinsburg, R. and Thomaz, C. (2020) “Neonatal Pain Assessment From Facial Expression Using Deep Neural Networks” In Anais do XVI Workshop de Visão Computacional (WVC 2020). Sociedade Brasileira de Computação - SBC. [link].

Carlini, L. P., Coutrin, G. de A. S., Ferreira, L. A., et al. (2024) “Human vs machine towards neonatal pain assessment: A comprehensive analysis of the facial features extracted by health professionals, parents, and convolutional neural networks” Artificial Intelligence in Medicine, v. 147, p. 102724.

Fontaine, D., Vielzeuf, V., Genestier, P., et al. (2022) “Artificial intelligence to evaluate postoperative pain based on facial expression recognition” European Journal of Pain, v. 26, n. 6, p. 1282–1291.

Guinsburg, R. e Cuenca A, M. C. C. (2019) “A linguagem da dor no recém-nascido” Documento Científico do Departamento de Neonatologia Sociedade Brasileira de Pediatria, v. 3, n. 9, p. 19–24.

Heiderich, T. M., Carlini, L. P., Buzuti, L. F., et al. (2023) “Face-based automatic pain assessment: challenges and perspectives in neonatal intensive care units” Jornal de Pediatria, v. 99, n. 6, p. 546–560.

Hoti, K., Chivers, P. T. and Hughes, J. D. (2021) “Assessing procedural pain in infants: a feasibility study evaluating a point-of-care mobile solution based on automated facial analysis” The Lancet Digital Health, v. 3, n. 10, p. e623–e634.

Macedo, J. S. e Müller, A. B. (2021) “Dor e medidas não-farmacológicas em prematuros hospitalizados” Revista Saúde - UNG-Ser, v. 15, n. 1/2, p. 23.

Othman, E., Werner, P., Saxen, F., et al. (2021) “Automatic vs. Human Recognition of Pain Intensity from Facial Expression on the X-ITE Pain Database” Sensors, v. 21, n. 9, p. 3273.

Salekin, M. S., Mouton, P. R., Zamzmi, G., et al. (2021) “Future roles of artificial intelligence in early pain management of newborns” Paediatric and Neonatal Pain, v. 3, n. 3, p. 134–145.

Silva, P. A. S. O. (2020) “Interpretação e reconhecimento de padrões para avaliação de dor em imagens faciais de recém-nascidos” Dissertação de Mestrado, Engenharia Elétrica, Centro Universitário FEI.

Zamzmi, G., Pai, C.-Y., Goldgof, D., et al. (2022) “A Comprehensive and ContextSensitive Neonatal Pain Assessment Using Computer Vision” IEEE Transactions on Affective Computing, v. 13, n. 1, p. 28–45.