سامانه کنگره های دانشگاه علوم پزشکی و خدمات بهداشتی و درمانی آذربایجان غربی

صفحه اصلی / بیستمین همایش سالیانه بیماری های شایع گوارش و کبد کودکان ایران و دومین همایش بین المللی چاقی کودکان

Artificial Intelligence for Early Detection and Prediction of Gastrointestinal Disorders in Children

نویسندگان :

Ladan Soltanzadeh¹ Ali Mirzaee AghGonbad² Shanli Mirzaee³

1- دانشگاه علوم علوم پزشکی ارومیه 2- دانشگاه تبریز 3- ارومیه

کلمات کلیدی :

Artificial Intelligence،Detection of Gastrointestinal Disorders،Prediction of Gastrointestinal Disorders،Children

چکیده :

Background and Aim: Childhood obesity is a global health crisis linked to type 2 diabetes, cardiovascular diseases, and psychosocial disorders, with rising prevalence in underserved and high-risk populations despite prevention efforts. Traditional interventions like nutritional counseling and exercise programs lack personalization and sustained efficacy. Emerging AI-powered wearables enable real-time tracking of physical activity, sleep, diet, and metabolic parameters, yet gaps persist in integrating multidimensional data and developing interpretable, scalable models for long-term impact. This review aims to analyze AI wearables and targeted interventions to establish an integrated framework for personalized, population-level strategies. Methods: A narrative systematic review was conducted, including studies published from 2020 to April 2025. Comprehensive searches in PubMed, EMBASE, Scopus, Cochrane Library, and EBSCO used keywords such as “artificial intelligence,” “wearable devices,” “childhood obesity,” “physical activity,” and “multidimensional monitoring.” Inclusion criteria encompassed original research, systematic reviews, and meta-analyses involving children and adolescents (0–18 years) that evaluated AI-powered or digital wearable interventions for obesity prevention or treatment. Results: Thirty-four studies were included. AI-driven wearables led to an average BMI reduction of 0.8–1.2 kg/m² over 6–12 months, with the greatest impact from personalized feedback systems. Moderate-to-vigorous physical activity increased by 15–25 minutes/day in 60% of studies. Machine learning models using wearable data achieved AUC values of 0.82–0.89 for obesity risk prediction, outperforming traditional criteria. Integration of physical, dietary, and physiological data improved prediction accuracy by 12–18%. However, only 18% of studies included underrepresented populations, and fewer than 25% assessed outcomes beyond 12 months. Conclusion: AI-powered wearables show strong potential for personalized, real-time childhood obesity prevention, improving BMI and activity outcomes. Integrating multidimensional data enhances prediction and intervention adherence, but challenges remain in scalability, equity, and long-term efficacy. Inclusive study designs and robust longitudinal trials are essential to translate these innovations into clinically actionable, equitable public health tools.