DeepHealthNet: Adolescent Obesity Prediction System Based on a Deep Learning Framework

Ji-Hoon Jeong, In-Gyu Lee, Sung-Kyung Kim, Tae-Eui Kam, Seong-Whan Lee, Euijong Lee

Abstract

Childhood and adolescent obesity rates are a global concern because obesity is associated with chronic diseases and long-term health risks. Artificial intelligence technology has emerged as a promising solution to accurately predict obesity rates and provide personalized feedback to adolescents. This study emphasizes the importance of early identification and prevention of obesity-related health issues. Factors such as height, weight, waist circumference, calorie intake, physical activity levels, and other relevant health information need to be considered for developing robust algorithms for obesity rate prediction and delivering personalized feedback. Hence, by collecting health datasets from 321 adolescents, we proposed an adolescent obesity prediction system that provides personalized predictions and assists individuals in making informed health decisions. Our proposed deep learning framework, DeepHealthNet, effectively trains the model using data augmentation techniques, even when daily health data are limited, resulting in improved prediction accuracy (acc: 0.8842). Additionally, the study revealed variations in the prediction of the obesity rate between boys (acc: 0.9320) and girls (acc: 0.9163), allowing the identification of disparities and the determination of the optimal time to provide feedback. The proposed system shows significant potential in effectively addressing childhood and adolescent obesity.

Introduction

Childhood and adolescent obesity is a global health crisis linked to long-term risks like cardiovascular disease and type 2 diabetes. While AI offers a path toward personalized preventative care, predicting weight trends in teenagers is notoriously difficult. The study utilized a custom-built dataset from 321 adolescents (aged 10–12) in Seoul, South Korea, tracking eight core metrics including BMI, step count, and calorie intake over an 8-week period.

The authors identify three primary hurdles in this domain:

  1. Data Scarcity and Quality: Daily health tracking is often inconsistent, leading to "holey" datasets with many missing values.
  2. Class Imbalance: In most studies, the majority of participants maintain their weight, while only a minority experience significant gain or loss. Standard models often fail to learn the patterns of these "minority" classes.
  3. Biological Variance: Factors influencing obesity manifest differently across genders and age groups, requiring models that can capture these nuances.

Methods

The researchers proposed DeepHealthNet, a deep learning architecture consisting of an input layer, three densely connected hidden layers (128, 256, and 128 nodes respectively), and an output layer for three classes: Increase, Maintenance, and Decrease.

DeepHealthNet model architecture and data preprocessing workflow.

Key technical innovations include:

  • Data Bundling: To combat missing daily records, the team used a "bundling" technique that averages $n$ days of data into single representative rows, preserving time-series characteristics while smoothing out noise.
  • SMOTE (Synthetic Minority Over-sampling Technique): To solve the class imbalance problem, they used SMOTE to mathematically generate synthetic examples of the "weight gain" and "weight loss" groups, ensuring the model didn't just default to predicting "weight maintenance."
  • WUDI! Ecosystem: Data was collected via a gamified app called "WUDI!" (Would U Do It!) which incentivized adolescents to track their habits using Samsung Galaxy Fit 2 smartwatches.

Key Results

DeepHealthNet achieved an overall accuracy of 0.8837, with a high F1-score of 0.8797. Paired t-tests confirmed that these improvements over Naïve Bayes, SVM, and LSTM were statistically significant ($p < 0.001$).

Performance comparison of DeepHealthNet against traditional machine learning models.
  • Model Comparison: DeepHealthNet's MLP architecture significantly outperformed the next best competitor, LSTM (0.7008 accuracy), proving that dense neural networks can be highly effective when paired with proper data augmentation.
  • Class-Specific Performance: The model was exceptionally good at identifying the "Decrease" class (99.05% True Positive Rate) and "Maintenance" (97.20%).
  • Gender-Based Findings: When split, the model achieved 0.9320 accuracy for boys and 0.9163 for girls. Interestingly, peak accuracy was reached earlier for girls (day 137) than boys (day 164), suggesting physical changes manifest more rapidly in girls during early adolescence.
  • Data Saturation: Performance reached a "saturation point" around the 60th day of collection, indicating that two months of consistent tracking is the threshold where the model truly begins to capture an individual's underlying health patterns.

Conclusion

DeepHealthNet demonstrates significant potential in addressing adolescent obesity through personalized predictions. While the study focused on a specific demographic in Seoul, it proves that dense neural networks, when paired with proper data augmentation like SMOTE and bundling, can overcome the challenges of sparse and imbalanced health data. Future work will look into more diverse demographics and additional variables like socioeconomic status and genetic predispositions to improve the generalizability of the system.

Summary updated

Abstract

Childhood and adolescent obesity rates are a global concern because obesity is associated with chronic diseases and long-term health risks. Artificial intelligence technology has emerged as a promising solution to accurately predict obesity rates and provide personalized feedback to adolescents. This study emphasizes the importance of early identification and prevention of obesity-related health issues. Factors such as height, weight, waist circumference, calorie intake, physical activity levels, and other relevant health information need to be considered for developing robust algorithms for obesity rate prediction and delivering personalized feedback. Hence, by collecting health datasets from 321 adolescents, we proposed an adolescent obesity prediction system that provides personalized predictions and assists individuals in making informed health decisions. Our proposed deep learning framework, DeepHealthNet, effectively trains the model using data augmentation techniques, even when daily health data are limited, resulting in improved prediction accuracy (acc: 0.8842). Additionally, the study revealed variations in the prediction of the obesity rate between boys (acc: 0.9320) and girls (acc: 0.9163), allowing the identification of disparities and the determination of the optimal time to provide feedback. The proposed system shows significant potential in effectively addressing childhood and adolescent obesity.
BibTeX
@article{jeong2024deephealthnet,
  author = {Ji-Hoon Jeong and In-Gyu Lee and Sung-Kyung Kim and Tae-Eui Kam and Seong-Whan Lee and Euijong Lee},
  title = {{DeepHealthNet: Adolescent Obesity Prediction System Based on a Deep Learning Framework}},
  journal = {IEEE J. Biomed. Health Informatics},
  year = {2024},
  url = {https://openreview.net/forum?id=cOejJLsA0I},
}
Figure