Motion Estimation and Characterization in Premature Newborns Using Long Duration Video Recordings - 08/08/17
, A. Simon a, b, M. Ugolin d, O. Rosec c, G. Carrault a, b, e, P. Pladys a, b, d, eAbstract |
Objectives |
In the context of neonatal non invasive monitoring, this paper proposes the estimation and characterization of the motion of premature newborns from long duration video recordings.
Material and Methods |
A set of 13 videos from 9 different patients, corresponding to 190 hours of recordings, have been studied. An algorithm based on the analysis of changes in the image border has been used to remove intervals artifacted by adults' presence. Then, some features were computed to characterize the baby's motion. The approach was applied to compare two groups of premature newborns, with different severities of prematurity, recorded at the same postmenstrual age.
Results |
Detection of adults' presence was achieved with 96.8% of sensitivity. All features were found statistically significant to differentiate the two groups.
Conclusion |
This study shows that the automated video monitoring on long periods is achievable and provides relevant information about the premature newborns motion activity.
Le texte complet de cet article est disponible en PDF.Graphical abstract |
Highlights |
• | Motion analysis has been shown to be relevant to observe premature newborns' behavior. |
• | Automated motion analysis including a method to detect adult's presence and a motion estimation method is proposed. |
• | A set of features is extracted from estimated motion series. |
• | Preliminary results indicate that the quantity of motion is greater for the most premature babies. |
Keywords : Premature newborn monitoring, Motion characterization, Video processing
Plan
Vol 38 - N° 4
P. 207-213 - août 2017 Retour au numéro
Article précédent
- Joint Watermarking and Lossless JPEG-LS Compression for Medical Image Security
- S. Haddad, G. Coatrieux, M. Cozic, D. Bouslimi
- Spectral CT Material Decomposition in the Presence of Poisson Noise: A Kullback–Leibler Approach
- T. Hohweiller, N. Ducros, F. Peyrin, B. Sixou
Bienvenue sur EM-consulte, la référence des professionnels de santé.
L’accès au texte intégral de cet article nécessite un abonnement.
Déjà abonné à cette revue ?