MOTOR OUTCOME & ASSESSMENT
Preterm birth is associated with increased risks of motor impairments such as cerebral palsy. The risks are highest in those born at the lowest gestation. Early identification of those most at risk is challenging meaning that a critical window of opportunity to improve outcomes through therapy-based interventions may be missed. We are combing the expertise of experienced neonatal physiotherapists (Claire Marcroft), paediatric neurologists (Dr Anna Basu) and computer scientists at Northumbria University (Dr Kevin McCay and Dr Edmund Ho) to develop technologies to better assess early patterns of movement.
Clinically, the assessment of spontaneous general movements ('Prechtl methodology') is an important tool, which can be used for the prediction of movement impairments. Movement recognition aims to capture and analyze relevant limb movements through computerized approaches focusing on continuous, objective, and quantitative assessment. Different methods of recording and analyzing infant movements have been explored including camera-based solutions, and miniaturized movement sensors. We explore machine learning methods and apply these to the analysis of the recorded movement data.
Preterm birth is associated with increased risks of motor impairments such as cerebral palsy. The risks are highest in those born at the lowest gestation. Early identification of those most at risk is challenging meaning that a critical window of opportunity to improve outcomes through therapy-based interventions may be missed. We are combing the expertise of experienced neonatal physiotherapists (Claire Marcroft), paediatric neurologists (Dr Anna Basu) and computer scientists at Northumbria University (Dr Kevin McCay and Dr Edmund Ho) to develop technologies to better assess early patterns of movement.
Clinically, the assessment of spontaneous general movements ('Prechtl methodology') is an important tool, which can be used for the prediction of movement impairments. Movement recognition aims to capture and analyze relevant limb movements through computerized approaches focusing on continuous, objective, and quantitative assessment. Different methods of recording and analyzing infant movements have been explored including camera-based solutions, and miniaturized movement sensors. We explore machine learning methods and apply these to the analysis of the recorded movement data.
SMART baby study
The early diagnosis of cerebral palsy has also been explored using tools such as the General Movements Assessment (GMA). We have explored the feasibility of extracting pose-based features from video sequences to automatically classify infant body movement into two categories, normal and abnormal. The classification was based upon the GMA, which was carried out on the video data by an independent expert reviewer. Dr McCay and Dr Ho are currently extending their previous work by extracting the normalised pose-based feature sets, Histograms of Joint Orientation 2D (HOJO2D) and Histograms of Joint Displacement 2D (HOJD2D), for use in new deep learning architectures.
The early diagnosis of cerebral palsy has also been explored using tools such as the General Movements Assessment (GMA). We have explored the feasibility of extracting pose-based features from video sequences to automatically classify infant body movement into two categories, normal and abnormal. The classification was based upon the GMA, which was carried out on the video data by an independent expert reviewer. Dr McCay and Dr Ho are currently extending their previous work by extracting the normalised pose-based feature sets, Histograms of Joint Orientation 2D (HOJO2D) and Histograms of Joint Displacement 2D (HOJD2D), for use in new deep learning architectures.
