Related Experiment Video
Updated: Aug 5, 2026
06:34
A Component-resolved Diagnostic Approach for a Study on Grass Pollen Allergens in Chinese Southerners with Allergic Rhinitis and/or Asthma
Published on: June 4, 2017
Trajectory Analysis of Pediatric Longitudinal Allergen Sensitization Patterns From Real World Data
Sungyun Kim1, Minjeong Lee2, Ji Soo Park3
1Interdisciplinary Program of Medical Informatics, Seoul National University College of Medicine, Seoul, Korea.
Allergy, Asthma & Immunology Research
|August 3, 2026
Summary
Early house dust mite sensitization predicts asthma development in children. Machine learning models analyzing allergen sensitization patterns can identify high-risk individuals for personalized allergy prevention strategies.
Area of Science:
- Computational immunology and allergy research.
- Application of machine learning in clinical diagnostics.
Background:
- Understanding allergen sensitization patterns is crucial for predicting allergic diseases.
- Longitudinal analysis of sensitization trajectories offers insights into disease development.
Purpose of the Study:
- To investigate the link between allergen sensitization patterns and allergic disease diagnoses.
- To identify static and dynamic sensitization clusters using unsupervised learning.
- To develop a predictive model for allergic disease based on early sensitization profiles.
Main Methods:
- Retrospective cohort study utilizing electronic health records from a common data model.
- Latent class analysis and Gaussian mixture models for cross-sectional and longitudinal sensitization clustering.
- Categorical Boosting multilabel classification with SHapley Additive exPlanations for predictive modeling.
Main Results:
- Four cross-sectional sensitization clusters identified: HDM-dominant, HDM + food, polysensitization, and multiple inhalant.
- Nine longitudinal trajectories revealed a shift from food to aeroallergen sensitization.
- Early house dust mite (HDM) sensitization strongly predicted asthma; eczema and food allergies linked to food/polysensitization trajectories.
Conclusions:
- Longitudinal sensitization trajectories, especially early HDM dominance, are predictive of asthma.
- Machine learning modeling of sensitization history aids early identification of high-risk children.
- Findings support personalized strategies for allergy prevention.
