Atlas of neonatal face images using triangular Meshes

Abstract

Over the last years, neonatal face analysis has allowed the investigation and creation of non-invasive methods that enable classification of painful stimulus in newborns. In this context, changes on facial movements and expression have provided relevant scientific information and clinical significance, since they describe the presence of the pain itself perceived by the newborns. In this work, we propose and implement a computational framework that uses triangular meshes, with the goal of generating high resolution spatially normalized atlases potentially useful for automatic neonatal pain assessment.

Face-based automatic pain assessment: challenges and perspectives in neonatal intensive care units

“…Other researchers have implemented a computational framework using triangular meshes to generate a spatially normalized atlas of high resolution, potentially useful for the automatic evaluation of neonatal pain. 19 These atlases are essential to describe characteristic and detailed facial patterns, preventing image effects or signals (which are not relevant and which portray undesirable particularities, inherent to the imperfect data acquisition process) from being erroneously propagated as discriminative variations.…”

“…The developed framework showed that it is possible to statistically classify the expression of pain and non-pain through facial images and highlight discriminant facial regions for the pain phenomenon. 19 …”

“…In the last years, computational methods have been developed to detect the painful phenomenon automatically, 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 to help health professionals to monitor the presence of pain and identify the need for therapeutic intervention. Even with the advancement of technology, these studies, all of them related to automatic neonatal pain assessment, have not addressed practical difficulties in identifying pain in newborns who remain with devices attached to their faces.…”

Interpretação e reconhecimento de padrões para avaliação de dor em imagens faciais de recém-nascidos

“…Para servir de entrada aos métodos de extração multivariada de informação, foram geradas novas imagens através da metodologia proposta por [Orona et al 2019], porém aplicando a política de todos-contra-todos, em vez de um-contra-todos. Para o um-contra-todos, utiliza-se uma imagem fixa de referência e todos as normalizações utilizam esta imagem como base, já o todos-contra-todos elimina a necessidade de se utilizar uma imagem de referência.…”