Grant: Master's Award: Canada Graduate Scholarships
University of Toronto
$17,500.00
Automated Video-based Analysis of the General Movements Assessment for Infants at Risk of Motor Impairment
EF 2017-18
Résumé
Preterm infants are at high risk of motor impairment with an incidence of cerebral palsy (CP) of 6.4% in those born 29 weeks gestational age (GA) and up to 40% in neonates born extremely preterm at 22-25 weeks GA. Early diagnosis is key in facilitating early targeted therapy, which has the potential to improve motor outcomes for these children in the future. The most sensitive screening tool is the General Movements Assessment (GMA), a visual assessment of an infant's movements performed by a trained expert. A recent systematic review of the literature noted that the GMA was the best available tool for early screening of cerebral palsy with a sensitivity and specificity of 98% and 91%, respectively. It is also non-invasive and cheaper than technologies such as Magnetic Resonance Imaging (MRI). However, it is not used routinely in Canada because of a lack of expertise and has limited generalizability because of its subjectivity. Automating the GMA is possible using two-dimensional (2D) video-based movement analysis and machine learning. This provides a quantitative alternative to visual assessment. Previous studies of automation have proved its feasibility, but clinical application has been limited by a lack of validation in larger groups of preterm infants. We propose a study that first uses an existing database of videos of preterm infants < 30 weeks GA to develop an automated video-analysis GMA tool with a clinically applicable score. This tool will then be validated in preterm infants for motor impairment at 18-24 months using prospectively collected videos. This technology can be implemented in neonatal follow-up clinics nationwide, including areas with less access to medical experts to reduce the costs and delays associated with diagnosis.
Programme
Master's Award: Canada Graduate Scholarships
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
CIHR Funding Decisions Database
Tel que publié
Id
386378
Name
Raghuram Kamini
Region
Ontario
Country
Canada
Orgtype
Institution Paid; Research Institution
Prcname2
Special Cases - Awards Programs
Projecttitle
Automated Video-based Analysis of the General Movements Assessment for Infants at Risk of Motor Impairment
Programtype2
Studentship Programs
Orgnameinp2
University of Toronto
Cihramount2
$17,500
Cihrequipment2
$0
Programname2
Master's Award: Canada Graduate Scholarships
Keyworddelim
Cerebral Palsy; Computerized General Movements Assessment; Early Intervention; Infants; Motor Skills Disorders; Movement; Neurodevelopmental Assessment; Prediction; Premature; Screening Tool; Translational Studies
Abstract
Preterm infants are at high risk of motor impairment with an incidence of cerebral palsy (CP) of 6.4% in those born 29 weeks gestational age (GA) and up to 40% in neonates born extremely preterm at 22-25 weeks GA. Early diagnosis is key in facilitating early targeted therapy, which has the potential to improve motor outcomes for these children in the future. The most sensitive screening tool is the General Movements Assessment (GMA), a visual assessment of an infant's movements performed by a trained expert. A recent systematic review of the literature noted that the GMA was the best available tool for early screening of cerebral palsy with a sensitivity and specificity of 98% and 91%, respectively. It is also non-invasive and cheaper than technologies such as Magnetic Resonance Imaging (MRI). However, it is not used routinely in Canada because of a lack of expertise and has limited generalizability because of its subjectivity. Automating the GMA is possible using two-dimensional (2D) video-based movement analysis and machine learning. This provides a quantitative alternative to visual assessment. Previous studies of automation have proved its feasibility, but clinical application has been limited by a lack of validation in larger groups of preterm infants. We propose a study that first uses an existing database of videos of preterm infants &lt; 30 weeks GA to develop an automated video-analysis GMA tool with a clinically applicable score. This tool will then be validated in preterm infants for motor impairment at 18-24 months using prospectively collected videos. This technology can be implemented in neonatal follow-up clinics nationwide, including areas with less access to medical experts to reduce the costs and delays associated with diagnosis.
Competitiondate
201712
Approvedterm2
1 yr 0 mth
Pinamesdelim
Raghuram, Kamini
Approvedterm
1 an 0 mois
Cihrcontribution2
~ $17,500~ $0~