Background: Mirror therapy (MT) creates a cerebral illusion of a normal movement in a paretic limb. Although mirror therapy has been studied as a suitable intervention for children with Unilateral Cerebral Palsy (UCP), a comprehensive understanding of its full range of benefits is still lacking. Thus, the aim of this systematic review and meta-analysis was to determine all motor and sensory effects of MT in children and adolescents with UCP. Methods: Clinical trials focused on the application of MT in the upper limb (UL) of children and adolescents with UCP were included. A search was performed in PubMed, Cochrane Library, Web of Science, and LILACS databases. Eleven studies were included in this systematic review. The PEDro scale and the MINORS scale were applied to evaluate the methodological quality of randomized and non-randomized controlled trials, respectively. The Risk of Bias tool was also employed to evaluate the potential bias. In addition, the TIDieR checklist was used to assess the quality of intervention reporting. A random-effects model was used for the meta-analysis. Results: The studies included children with UCP from three to eighteen years, classified in Manual Ability Classification System levels I–IV. Motor effects of MT were found in nine studies. Also, two studies reported sensory effects on registration, perception, and proprioception abilities. Qualitative and quantitative analysis showed that MT improved manual dexterity and tactile registration in children and adolescents with UCP. Conclusions: MT is a therapy capable of inducing motor and sensory improvements in the affected UL of children with UCP.
Background: Artificial intelligence (AI)-driven markerless motion capture (MMC) technologies are increasingly being integrated into pediatric healthcare to improve the assessment and management of movement disorders. These video-based systems enable non-invasive motion analysis without wearable sensors, facilitating more natural movement assessment in children, particularly those with neurological or developmental conditions. Objectives: We evaluated the clinical applicability of AI-based MMC tools in pediatric settings for diagnosis, monitoring of motor development, and rehabilitation. Methods: This systematic review was registered in PROSPERO (CRD42024511787) and conducted by two independent reviewers, with a third reviewer resolving disagreements. The literature published between 2018 and 2025 was systematically searched. Studies involving pediatric populations or clinically relevant pediatric applications of MMC were included. Results: Of 1521 identified studies, 52 were finally selected. The included studies evaluated populations across a wide age range. However, seven of the included articles were specifically focused on underage populations. Infant studies primarily analyzed whole-body movements, emphasizing the relevance of global motor patterns in early development. OpenPose and AlphaPose were the most frequently used frameworks in pediatric research because of their automatic full-body key point detection, whereas DeepLabCut was commonly selected for its customizable labeling capabilities. Theia3D emerged as a promising clinically applicable solution with high accuracy. Most studies evaluated kinematic parameters as objective markers of motor performance and development. However, methodological heterogeneity and limited pediatric-specific validation remain important limitations. Conclusions: AI-driven MMC technologies show considerable potential to support objective, accessible, and child-friendly movement assessment in pediatric clinical practice.