Artificial Intelligence in Participation-Focused Pediatric Rehabilitation: A Scoping Review

Objective(s) To examine how artificial intelligence (AI) is integrated into pediatric rehabilitation interventions targeting the participation of children and youth with childhood disabilities in valued activities. Data Sources For this scoping review, we conducted a systematic literature search in well-established applied health sciences and computer science databases for articles published before February 2020. Study Selection Two independent researchers screened and selected studies based on the following inclusion criteria: 1) participation was an explicit study aim or outcome, or the targeted focus of the AI application; 2) AI was applied as part of the provided and tested intervention; 3) children or youth with a childhood disability were the focus of the study and/or AI application; and 4) the article was published in English. Data Extraction Two researches mapped the data according to types of AI, mode of delivery, type of personalization, and whether the intervention addressed individual goal-setting. Data Synthesis The literature search identified 1,852 articles of which 63 met the inclusion criteria. Most included articles used multiple AI applications with the highest prevalence of robotics (87%) and human-agent/computer/robot interaction (56%). As for mode of delivery, most of the included articles described an intervention delivered in-person (89%) and only 7 (11%) were delivered remotely. Only one was customized to patients’ individual needs (2%). No intervention described individual goal setting as part of their therapy process or intervention planning. Conclusions Three major gaps for further research and development were identified: 1) A lack of remotely provided participation-focused interventions using AI; 2) a lack of individual goal-setting integrated in pediatric rehabilitation interventions; and 3) a lack of rehabilitation interventions tailored to the individual needs of children, youth and/or families. Author(s) Disclosures We have no conflict of interest to declare. To examine how artificial intelligence (AI) is integrated into pediatric rehabilitation interventions targeting the participation of children and youth with childhood disabilities in valued activities. For this scoping review, we conducted a systematic literature search in well-established applied health sciences and computer science databases for articles published before February 2020. Two independent researchers screened and selected studies based on the following inclusion criteria: 1) participation was an explicit study aim or outcome, or the targeted focus of the AI application; 2) AI was applied as part of the provided and tested intervention; 3) children or youth with a childhood disability were the focus of the study and/or AI application; and 4) the article was published in English. Two researches mapped the data according to types of AI, mode of delivery, type of personalization, and whether the intervention addressed individual goal-setting. The literature search identified 1,852 articles of which 63 met the inclusion criteria. Most included articles used multiple AI applications with the highest prevalence of robotics (87%) and human-agent/computer/robot interaction (56%). As for mode of delivery, most of the included articles described an intervention delivered in-person (89%) and only 7 (11%) were delivered remotely. Only one was customized to patients’ individual needs (2%). No intervention described individual goal setting as part of their therapy process or intervention planning. Three major gaps for further research and development were identified: 1) A lack of remotely provided participation-focused interventions using AI; 2) a lack of individual goal-setting integrated in pediatric rehabilitation interventions; and 3) a lack of rehabilitation interventions tailored to the individual needs of children, youth and/or families.