Is Smiling the Key? Machine Learning Analytics Detect Subtle Patterns in Micro-Expressions of Infants with ASD

Time is a key factor to consider in Autism Spectrum Disorder. Detecting the condition as early as possible is crucial in terms of treatment success. Despite advances in the literature, it is still difficult to identify early markers able to effectively forecast the manifestation of symptoms. Artificial intelligence (AI) provides effective alternatives for behavior screening. To this end, we investigated facial expressions in 18 autistic and 15 typical infants during their first ecological interactions, between 6 and 12 months of age. We employed Openface, an AI-based software designed to systematically analyze facial micro-movements in images in order to extract the subtle dynamics of Social Smiles in unconstrained Home Videos. Reduced frequency and activation intensity of Social Smiles was computed for children with autism. Machine Learning models enabled us to map facial behavior consistently, exposing early differences hardly detectable by non-expert naked eye. This outcome contributes to enhancing the potential of AI as a supportive tool for the clinical framework.

Randomized Controlled Trial of Electronic Health Risk Behavior Screening with Integrated Feedback among Adolescents in Primary Care (Preprint)

BACKGROUND Health risk behaviors are the most common sources of morbidity among adolescents. Guidelines recommend screening and counseling but implementation is inconsistent. OBJECTIVE To test the efficacy of electronic risk behavior screening with integrated patient-facing feedback on the delivery of adolescent-reported clinician counseling and risk behaviors over time. METHODS Randomized controlled trial comparing the electronic tool with usual care. Outcomes assessed via online survey at 1-day and 3- and 6-months post visit. 300 13-18 year olds attending a well-care visit between September 30, 2016 and January 12, 2018. Adolescents were randomized after consent employing a 1:1 balanced age, sex, and clinic stratified schema with 150 adolescents in the intervention group and 150 in the control group. Of the original sample, 282 (94%), 283 (94%), and 284 (95%) completed follow-up surveys at 1-day, 3-months, and 6-months respectively with similar levels of attrition across study arms. Intervention adolescents received electronic screening and feedback and clinicians received a summary report of results. Control adolescents received usual care. RESULTS The mean risk behavior score at baseline was 2.86 (SD 2.33) for intervention adolescents and 3.10 (SD 2.52) for control adolescents (Score potential range: 0 to 20) After adjusting for age, gender, and random effect of clinic, intervention adolescents were 36% more likely to report having received counseling for endorsed risk behaviors than control adolescents (aRR-1.36, 95% CI: 1.04,1.78) at 1-day post well-care visit. Both the intervention group and the control group reported decreased risk behaviors at 3- and 6-month follow-up assessments with no significant group differences in risk behavior scores at either time point. CONCLUSIONS While electronic health screening with integrated feedback improves the delivery of counseling by clinicians, the impact on risk behaviors is modest and, in this study, not significantly different than usual care. More research is needed to identify effective strategies to reduce risk in the context of well care. CLINICALTRIAL clinicaltrials.gov Identifier: NCT02882919

DEVELOPING SYMPTOMATIC BEHAVIOR SCREENING TOOL (SYMBEST) FOR CHILDREN WITH BEHAVIOR PROBLEMS : A FUZZY DELPHI METHOD

The purpose of this research is to develop a symptomatic behaviour screening tool (SymBest) for early childhood educators to identify children with symptomatic behaviours. The measuring constructs of the screening tool are the child developmental domains with developmental delays as items representing the constructs. Fuzzy Delphi analysis was conducted with 18 participants from diverse backgrounds of clinical and education to gain the expert consensus on the suitability of the constructs and items representing SymBest. The findings showed that the experts have a fair degree of agreement on the constructs and the items suggested to form SymBest. The constructs and items with accepted threshold value, percentage of group consensus and fuzzy score is then organized in sequence priority to form the screening tool.

Measurement of Salivary pH Using Litmus Paper as a Screening Tool of Smoking Behavior in Adolescents

Background: Cigarettes are the biggest threat in the world, including Indonesia. According to The Health Ministry of Indonesia 2017 , Indonesia is one the country with the highest number of smokers. More than one-third (36.3%) of Indonesia's population, 20% are adolescents aged 13-15 years old. In Indonesia, cigarettes caused more than 200,000 deaths per year. It has been recorded that the amount of cigarettes consumed in Indonesia increases from 182 billion in 2001 to 260.8 billion in 2009. In addition to the increasing number of the smokers and the age of smoking, there is another factor which is adolescents deceit for their smoking behavior, therefore screening is necessary to prevent the complications from smoking. Aim: To find out changes in salivary pH as a smoking behavior screening tool in adolescents. Methods: : Results: Distribution of data in 85 samples with 50 nonsmokers and 35 smokers, got the youngest age of smoking is in fourth year of elementary school, with the highest number of cigarettes consumed per day is more than 2 boxes. The salivary pH in smokers was significantly ( P < 0.005) lower than nonsmokers. This can be seen also through since the smoke, the last time smoking and the number of cigarettes consumed per day. The younger age of smoking, the last time smoking (<24 hour) and the more the number of cigarettes consumed per day, that all can significantly decrease salivary pH. Conclusion: Salivary pH measurement using litmus paper is a potential screening for smoking behavior in adolescents.