Identification of obesity risk factors in 3-12-year-old children and adolescents with prior respiratory tract infections via interpretable machine and deep learning models
Identifikation von Adipositas-Risikofaktoren bei 3- bis 12-jährigen Kindern und Jugendlichen mit früheren Atemwegsinfektionen mittels interpretierbarer Machine-Learning- und Deep-Learning-Modelle
JAMIA Open · 07.07.2026 · 9 (4)
DOI: 10.1093/jamiaopen/ooag061 ↗Auf einen Blick
Bei 6.509 Kindern und Jugendlichen mit früheren RTIs sagte ein LightGBM-Modell Adipositas am besten voraus. 20 Schlüsselmerkmale wurden identifiziert; ein Sequenznetzwerk bestätigte deren Vorhersagewert.
Fragestellung
Welche Faktoren sagen Adipositas bei 3- bis 12-jährigen Kindern und Jugendlichen mit früheren RTIs voraus?
Methode
Querschnittstudie mit 6.509 Kindern und Jugendlichen aus Beijing und Tangshan; 12 ML-Modelle zur Vorhersage von Adipositas vs. Normalgewicht; SHAP und Validierung mit Sequenznetzwerk.
Ergebnisse
LightGBM erzielte die beste Leistung (Accuracy 0,8844; AUC 0,9491). Das Sequenznetzwerk bestätigte die Merkmale mit Accuracy 0,8023 und AUC 0,8117.
Schlussfolgerung der Autor:innen
Das LightGBM-basierte Modell ermöglicht eine effektive Vorhersage des Adipositasrisikos bei Kindern mit früheren RTIs; die identifizierten Merkmale können frühes Screening und personalisierte Interventionen unterstützen.
Kernaussagen
- LightGBM war das beste der 12 ML-Modelle
- 20 Schlüsselmerkmale wurden per SHAP identifiziert
- Ein Sequenznetzwerk bestätigte den Vorhersagewert der Merkmale
Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.
Kurz gefasst
Fachartikel in der Zeitschrift JAMIA Open, online erschienen am 06.08.2026. Verfasst von Xiao-Qian Wang und 6 weiteren Autor:innen. Beteiligte Einrichtung: Capital Institute of Pediatrics, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing 100020, China. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, COVID-19 Pandemic on Child Growth and Development in China”), the Public Service Development and Reform Pilot.
Themen: Kinderkrankenpflege
Abstract (Originalsprache)
Backgrounds and Objectives Childhood obesity and respiratory tract infections (RTIs) are 2 major global public health issues that frequently co-occur and are closely interrelated. Early detection of children with prior RTIs who are at high obesity risk is crucial for targeted interventions. This study integrates interpretable machine learning (ML) models and a deep learning network to develop an obesity risk prediction model in a large pediatric cohort.
Methods
Cross-sectional data from 6509 children and adolescents aged 3-12 years with prior RTIs in Beijing and Tangshan were fed to 12 ML models to predict childhood obesity (versus children with normal weight). Bayesian optimization was applied to fine-tune model hyperparameters. Prediction performance was assessed using 8 metrics. Key predictive features were identified by SHapley Additive exPlanations (SHAP). The validity of the optimal ML model was verified by the sequential neural network model.
Results
Of 12 ML models, LightGBM achieved the optimal performance (accuracy: 0.8844, area under the curve [AUC]: 0.9491). SHAP analysis identified 20 key predictors, including child age, paternal body mass index (BMI), maternal BMI, birth length, birthweight, gestational age, eating speed, complementary feeding initiation age, screen time, breastfeeding duration, bedtime, maternal age, nighttime sleep, outdoor activity, dental caries, sedentary time, food allergies, family history of diabetes, and sex. The deep learning sequence network model further validated the predictive value of these features (accuracy: 0.8023, AUC: 0.8117), and the SHAP-driven feature importance rankings were in close line with LightGBM.
Conclusions
Our LightGBM-based model enables effective prediction of obesity risk in children aged 3-12 years with prior RTIs, and the key features identified can inform early screening and facilitate personalized interventions.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Capital Institute of Pediatrics, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing 100020, China.
- Center for Evidence-Based Medicine, Capital Center for Children's Health, Capital Medical University, Capital Institute of Pediatrics, Beijing 100020, China.
- Graduate School, Beijing University of Chinese Medicine, Beijing 100029, China.
- Department of Pediatrics, China-Japan Friendship Hospital, Beijing 100029, China.
- Institute of Clinical Medical Sciences, China-Japan Friendship Hospital, Beijing 100029, China.
ORCID-Profile: Wen-Quan Niu
Förderung
- National Natural Science Foundation of China
- COVID-19 Pandemic on Child Growth and Development in China”)
- the Public Service Development and Reform Pilot
- the Capital’s Funds for Health Improvement and Research
Zitieren
APA
Wang, X. Q., Zheng, F. J., Wang, Q., Li, C., Zhang, W. Q., Zhang, Z. X., & Niu, W. Q. (2026). Identification of obesity risk factors in 3-12-year-old children and adolescents with prior respiratory tract infections via interpretable machine and deep learning models. JAMIA Open, 9(4). https://doi.org/10.1093/jamiaopen/ooag061
Vancouver
Wang XQ, Zheng FJ, Wang Q, Li C, Zhang WQ, Zhang ZX, et al. Identification of obesity risk factors in 3-12-year-old children and adolescents with prior respiratory tract infections via interpretable machine and deep learning models. JAMIA Open. 2026;9(4). doi:10.1093/jamiaopen/ooag061
BibTeX
@article{Wang2026,
author = {Wang, Xiao-Qian and Zheng, Fang-Jie-Yi and Wang, Qiong and Li, Che and Zhang, Wen-Qian and Zhang, Zhi-Xin and Niu, Wen-Quan},
title = {Identification of obesity risk factors in 3-12-year-old children and adolescents with prior respiratory tract infections via interpretable machine and deep learning models},
journal = {JAMIA Open},
year = {2026},
volume = {9},
number = {4},
doi = {10.1093/jamiaopen/ooag061}
}
EUVIMED