Results of the spontaneous movement assessment project using the SENDD (system for the early detection of neurological abnormalities) in infants

Assessment of spontaneous movements (General Motion Assessment - GMA) has been proven to be a reliable method for identifying neurological impairments in early infancy. Despite the fact that it does not use complicated technology (it is a visual gestalt analysis of videos), the implementation is difficult because it requires time for the training of the examiner and for the analysis of the video of an individual patient, thus consuming the resources of the healthcare system. In a period of increasingly intensive use of artificial intelligence in medicine, the goal is to enable the automated assessment of spontaneous movements recorded in home conditions using a system of machine learning and computer vision, using the system for early neurologic deviation detection (SENDD) in infants. The results of the application of SENDD so far support its use as a screening tool which can allow professionals more time and resources to deal with those children who exhibit deviations from normal spontaneous movements.

Key words:
artificial intelligence; general movements assessment; machine learning; neural networks; neurological development