A Machine Learning Strategy for Autism Screening in Toddlers.

OBJECTIVE: Autism spectrum disorder (ASD) screening can improve prognosis via early diagnosis and intervention, but lack of time and training can deter pediatric screening. The Modified Checklist for Autism in Toddlers, Revised (M-CHAT-R) is a widely used screener but requires follow-up questions and error-prone human scoring and interpretation. We consider an automated machine learning (ML) method for overcoming barriers to ASD screening, specifically using the feedforward neural network (fNN). METHODS: The fNN technique was applied using archival M-CHAT-R data of 14,995 toddlers (age 16-30 months, 46.51% male). The 20 M-CHAT-R items were inputs, and ASD diagnosis after follow-up and diagnostic evaluation (i.e., ASD or not ASD) was the output. The sample was divided into subgroups by race (i.e., white and black), sex (i.e., boys and girls), and maternal education (i.e., below and above 15 years of education completed) to examine subgroup differences. Each subgroup was evaluated for best-performing fNN models. RESULTS: For the total sample, best results yielded 99.72% correct classification using 18 items. Best results yielded 99.92% correct classification using 14 items for white toddlers and 99.79% correct classification using 18 items for black toddlers. In boys, best results yielded 99.64% correct classification using 18 items, whereas best results yielded 99.95% correct classification using 18 items in girls. For the case when maternal education is 15 years or less (i.e., associate degree and below), best results were 99.75% correct classification when using 16 items. Results were essentially the same when maternal education was 16 years or more (i.e., above associate degree); that is, 99.70% correct classification was obtained using 16 items. CONCLUSION: The ML method was comparable to the M-CHAT-R with follow-up items in accuracy of ASD diagnosis while using fewer items. Therefore, ML may be a beneficial tool in implementing automatic, efficient scoring that negates the need for labor-intensive follow-up and circumvents human error, providing an advantage over previous screening methods.

Examining the Feasibility of Smartphone Game Applications for Physical Activity Promotion in Middle School Students.

OBJECTIVE: The objective of this study was to determine the feasibility (i.e., limited efficacy testing, practicality, and acceptability) of a 6-week smartphone game-based applications program for promoting physical activity (PA) in adolescents in an afterschool program. MATERIALS AND METHODS: This mixed-method, quasi-experimental design study included 27 adolescents who evaluated four smartphone PA game-based applications in two Boys & Girls Clubs of America. After an initial baseline week (i.e., usual activity during their visit to the Club), adolescents played each game for 1 week. During a final week, the participants could choose to play any combination of the four games. An established conceptual framework was used to assess feasibility. Efficacy was assessed by changes in PA via wrist-worn accelerometers (model GT3x+; ActiGraph LLC, Pensacola, FL). Practicality was measured through field notes, the number of players attending each session, and the proportion of attendees who played the games. Acceptability was measured using poststudy focus groups. RESULTS: Compared with baseline (3.22 metabolic equivalents [METs]), mean accelerometer values were significantly (P<0.05) higher during "Space Rayders" (4.33 METs) and "Color Hunt" (3.67 METs). Attendance did not differ among games, and weekly number of players averaged 12 of 27 participants. Qualitative findings indicated that participants perceived "Space Rayders" as the most acceptable game. Overall, participants found the games to be enjoyable and easy to use, although they had suggestions to improve graphics and sounds. CONCLUSIONS: Smartphone games can be feasible for adolescents to use for PA. Lessons learned will be used to provide improvements for future game development and evaluation.