This study examined multiple variables obtained from an automated measure of lip movement during a diadochokinesis (DDK) task to identify those with potential to detect mild speech motor involvement in school-age children diagnosed with cerebral palsy (CP).

Eight children with CP and high speech intelligibility and a matched group of eight children with typical development (TD) completed a DDK task while their lip and jaw movements were recorded. A custom MATLAB algorithm was used to automatically extract 23 kinematic measures of children's lip movements during production of the DDK sequences. Mann–Whitney U tests were used to compare groups on the kinematic measures, and receiver operating characteristic (ROC) analysis was used to evaluate the diagnostic accuracy of measures that significantly differed between groups.

Five of the 23 kinematic variables differed significantly between the CP and TD groups. These were two measures of overall DDK performance (i.e., duration of the DDK sequence and number of cycles) and three spatial and temporal measures of lip movement. Duration of the DDK sequence and the mean displacement of the lips across cycles had the highest diagnostic accuracy, differentiating CP and TD groups with 88% sensitivity and 88% specificity.

Automatically derived kinematic measures of DDK sequences differentiated children with CP and high intelligibility from typically developing children. Future research is needed to determine the clinical utility of these measures for detecting speech motor impairment.