Title
A method to evaluate skill transfer and acquisition of obstetric gestures based on the curvatures analysis of the position and the orientation.
Abstract
This paper focuses on the gesture analysis in order to compare two human gestures. The orientations and the positions of the gestures are both taken into account and the similarity rate between two gestures is calculated. In our case, the application is in obstetrics and the aim is to evaluate forceps blade placement. The method is based on the curvature analysis of the paths during the gesture. The 3-D position paths are expressed according to their cumulated chord length and the orientation paths in the quaternion unit space. These parameterizations lead to analyze data in space independently to time as requested by physicians. After filtering data in order to minimize sensor noises, the gestures are then compared by calculating the correlation between the position and the orientation curvatures of a novice gesture and an expert one. The results clearly show that novice skills in handling forceps increase in becoming smoother and closer to the reference placement. A childbirth simulator allows novices to acquire experience without any risks, however the training have to be completed with the extraction gesture evaluation and a compulsory training period in the delivery ward.
Year
DOI
Venue
2008
10.1016/j.jbi.2008.03.012
Journal of Biomedical Informatics
Keywords
Field
DocType
human gesture,obstetric gesture,forceps increase,curvatures analysis,quaternions,novice skill,skill transfer,training,novice gesture,curvature analysis,gesture analysis,compulsory training period,childbirth simulator,3-d position path,skill evaluation,forceps blade placement,extraction gesture evaluation,learning,cumulant
Computer vision,Gesture,Skill transfer,Computer science,Quaternion,Filter (signal processing),Gesture analysis,Correlation,Artificial intelligence,Chord (geometry),Curvature analysis
Journal
Volume
Issue
ISSN
41
6
1532-0480
Citations 
PageRank 
References 
4
0.82
11
Authors
6
Name
Order
Citations
PageRank
Richard Moreau1124.60
V Ochoa240.82
M T Pham340.82
P Boulanger440.82
T. Redarce5546.13
Olivier Dupuis681.72