OBJECTIVE: Absence of fidgety movements (FM) in high-risk infants is a strong marker for later cerebral palsy (CP). FMs can be classified by the General Movement Assessment (GMA), based on Gestalt perception of the infant's movement pattern. More objective movement analysis may be provided by computer-based technology. The aim of this study was to explore the feasibility of a computer-based video analysis of infants' spontaneous movements in classifying non-fidgety versus fidgety movements. METHOD: GMA was performed from video material of the fidgety period in 82 term and preterm infants at low and high risks of developing CP. The same videos were analysed using the developed software called General Movement Toolbox (GMT) with visualisation of the infant's movements for qualitative analyses. Variables derived from the calculation of displacement of pixels from one video frame to the next were used for quantitative analyses. RESULTS: Visual representations from GMT showed easily recognisable patterns of FMs. Of the eight quantitative variables derived, the variability in displacement of a spatial centre of active pixels in the image had the highest sensitivity (81.5) and specificity (70.0) in classifying FMs. By setting triage thresholds at 90% sensitivity and specificity for FM, the need for further referral was reduced by 70%. CONCLUSION: Video recordings can be used for qualitative and quantitative analyses of FMs provided by GMT. GMT is easy to implement in clinical practice, and may provide assistance in detecting infants without FMs.
OBJECTIVE: Absence of fidgety movements (FM) in high-risk infants is a strong marker for later cerebral palsy (CP). FMs can be classified by the General Movement Assessment (GMA), based on Gestalt perception of the infant's movement pattern. More objective movement analysis may be provided by computer-based technology. The aim of this study was to explore the feasibility of a computer-based video analysis of infants' spontaneous movements in classifying non-fidgety versus fidgety movements. METHOD: GMA was performed from video material of the fidgety period in 82 term and preterm infants at low and high risks of developing CP. The same videos were analysed using the developed software called General Movement Toolbox (GMT) with visualisation of the infant's movements for qualitative analyses. Variables derived from the calculation of displacement of pixels from one video frame to the next were used for quantitative analyses. RESULTS: Visual representations from GMT showed easily recognisable patterns of FMs. Of the eight quantitative variables derived, the variability in displacement of a spatial centre of active pixels in the image had the highest sensitivity (81.5) and specificity (70.0) in classifying FMs. By setting triage thresholds at 90% sensitivity and specificity for FM, the need for further referral was reduced by 70%. CONCLUSION: Video recordings can be used for qualitative and quantitative analyses of FMs provided by GMT. GMT is easy to implement in clinical practice, and may provide assistance in detecting infants without FMs.
Authors: Ragnhild Støen; Nils Thomas Songstad; Inger Elisabeth Silberg; Toril Fjørtoft; Alexander Refsum Jensenius; Lars Adde Journal: Pediatr Res Date: 2017-07-26 Impact factor: 3.756
Authors: Lisa Ouss; Marie-Thérèse Le Normand; Kevin Bailly; Marluce Leitgel Gille; Christelle Gosme; Roberta Simas; Julia Wenke; Xavier Jeudon; Stéphanie Thepot; Telma Da Silva; Xavier Clady; Edith Thoueille; Mohammad Afshar; Bernard Golse; Mariana Guergova-Kuras Journal: Front Psychol Date: 2018-02-19
Authors: Christa Einspieler; Peter B Marschik; Jasmin Pansy; Anna Scheuchenegger; Magdalena Krieber; Hong Yang; Maria K Kornacka; Edyta Rowinska; Marina Soloveichick; Arend F Bos Journal: Dev Med Child Neurol Date: 2015-09-14 Impact factor: 5.449
Authors: Muhammad Tausif Irshad; Muhammad Adeel Nisar; Philip Gouverneur; Marion Rapp; Marcin Grzegorzek Journal: Sensors (Basel) Date: 2020-09-17 Impact factor: 3.576