“…Most of the 94 included studies used robotics as the type of AI intervention to target participation among children and youth with disabilities or other diagnosed health conditions (72/94, 77%) [ 35 - 106 ], followed by HMI (51/94, 54%) [ 35 , 37 - 44 , 47 , 50 - 55 , 58 , 65 , 66 , 69 , 71 , 72 , 75 , 78 , 80 , 82 , 85 - 89 , 91 - 98 , 100 , 101 , 103 - 111 ], visualization and VR (19/94, 20%) [ 53 , 54 , 72 , 79 , 107 , 108 , 112 - 124 ], NLP (18/94, 19%) [ 36 , 47 , 52 , 64 , 71 , 78 , 79 , 82 , 91 , 101 , 103 - 105 , 120 , 125 - 128 ], ML (11/94, 12%), computer vision (10/94, 11%) [ 40 , 41 , 64 , 67 , 69 , 107 , 110 , 120 , 125 , 126 , 128 ], and constraint satisfaction and optimization (1/94, 1%; Table 1 ) [ 110 ]. Of the 72 studies on robotics, 63 (88%) studies...…”