Comparative Analysis of Binary and Multiclass Activity Recognition in High-Quality Newborn Resuscitation Videos

Jorge García-Torres, Øyvind Meinich-Bache, Siren Irene Rettedal, Amalie Kibsgaard, Sara Brunner, Kjersti Engan
Proceedings of the 5th Northern Lights Deep Learning Conference ({NLDL}), PMLR 233:59-66, 2024.

Abstract

Globally, 3-10% of newborns do not breathe spontaneously at birth and need resuscitation. Prompt initiation of resuscitative interventions such as tactile stimulation and positive pressure ventilation can reduce neonatal mortality and morbidity associated with birth asphyxia. Automated video analysis of resuscitation episodes may be beneficial for evaluation and debriefing purposes. In this work, a dataset of 220 newborn resuscitation videos collected at the Stavanger University Hospital (Norway) is used to develop NBT-I3D, a deep neural network pipeline to automatically recognize resuscitation activities. To assess the task, both binary and multiclass networks have undergone training, allowing for a comparison of the two approaches. Results obtained for binary classification show a mean precision and recall of 84.76% and 80.92%, respectively. For multiclass, a mean precision and recall of 72.26% and 74.80% are reported.

Cite this Paper

BibTeX
@InProceedings{pmlr-v233-garcia-torres24a, title = {Comparative Analysis of Binary and Multiclass Activity Recognition in High-Quality Newborn Resuscitation Videos}, author = {Garc\'ia-Torres, Jorge and Meinich-Bache, {\O}yvind and Rettedal, Siren Irene and Kibsgaard, Amalie and Brunner, Sara and Engan, Kjersti}, booktitle = {Proceedings of the 5th Northern Lights Deep Learning Conference ({NLDL})}, pages = {59--66}, year = {2024}, editor = {Lutchyn, Tetiana and Ramírez Rivera, Adín and Ricaud, Benjamin}, volume = {233}, series = {Proceedings of Machine Learning Research}, month = {09--11 Jan}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v233/garcia-torres24a/garcia-torres24a.pdf}, url = {https://proceedings.mlr.press/v233/garcia-torres24a.html}, abstract = {Globally, 3-10% of newborns do not breathe spontaneously at birth and need resuscitation. Prompt initiation of resuscitative interventions such as tactile stimulation and positive pressure ventilation can reduce neonatal mortality and morbidity associated with birth asphyxia. Automated video analysis of resuscitation episodes may be beneficial for evaluation and debriefing purposes. In this work, a dataset of 220 newborn resuscitation videos collected at the Stavanger University Hospital (Norway) is used to develop NBT-I3D, a deep neural network pipeline to automatically recognize resuscitation activities. To assess the task, both binary and multiclass networks have undergone training, allowing for a comparison of the two approaches. Results obtained for binary classification show a mean precision and recall of 84.76% and 80.92%, respectively. For multiclass, a mean precision and recall of 72.26% and 74.80% are reported.} }
Endnote
%0 Conference Paper %T Comparative Analysis of Binary and Multiclass Activity Recognition in High-Quality Newborn Resuscitation Videos %A Jorge García-Torres %A Øyvind Meinich-Bache %A Siren Irene Rettedal %A Amalie Kibsgaard %A Sara Brunner %A Kjersti Engan %B Proceedings of the 5th Northern Lights Deep Learning Conference ({NLDL}) %C Proceedings of Machine Learning Research %D 2024 %E Tetiana Lutchyn %E Adín Ramírez Rivera %E Benjamin Ricaud %F pmlr-v233-garcia-torres24a %I PMLR %P 59--66 %U https://proceedings.mlr.press/v233/garcia-torres24a.html %V 233 %X Globally, 3-10% of newborns do not breathe spontaneously at birth and need resuscitation. Prompt initiation of resuscitative interventions such as tactile stimulation and positive pressure ventilation can reduce neonatal mortality and morbidity associated with birth asphyxia. Automated video analysis of resuscitation episodes may be beneficial for evaluation and debriefing purposes. In this work, a dataset of 220 newborn resuscitation videos collected at the Stavanger University Hospital (Norway) is used to develop NBT-I3D, a deep neural network pipeline to automatically recognize resuscitation activities. To assess the task, both binary and multiclass networks have undergone training, allowing for a comparison of the two approaches. Results obtained for binary classification show a mean precision and recall of 84.76% and 80.92%, respectively. For multiclass, a mean precision and recall of 72.26% and 74.80% are reported.
APA
García-Torres, J., Meinich-Bache, Ø., Rettedal, S.I., Kibsgaard, A., Brunner, S. & Engan, K.. (2024). Comparative Analysis of Binary and Multiclass Activity Recognition in High-Quality Newborn Resuscitation Videos. Proceedings of the 5th Northern Lights Deep Learning Conference ({NLDL}), in Proceedings of Machine Learning Research 233:59-66 Available from https://proceedings.mlr.press/v233/garcia-torres24a.html.

Related Material