Multimodal Tongue-Pulse Information Fusion for Syndrome Diagnosis and Cohort Study in Children With Asthma

Overview

Asthma is one of the most prevalent chronic respiratory diseases in children, and accurate phenotyping and disease monitoring remain challenging in routine clinical practice. This observational cohort study aims to investigate the clinical value of multimodal tongue and pulse information in the syndrome diagnosis and phenotypic characterization of pediatric asthma. Children aged 5-18 years with a confirmed diagnosis of asthma will be enrolled at Shanghai Children's Medical Center and followed in routine outpatient care. Standardized tongue images and pulse wave data will be collected using validated acquisition devices during visits when lung function testing is performed. Quantitative features extracted from tongue and pulse data will be integrated with clinical information, including asthma stage, lung function parameters, eosinophil counts, allergic sensitization status, and Asthma Control Questionnaire-5 (ACQ-5) scores. The primary objective is to evaluate the associations between tongue-pulse multimodal features and asthma clinical stages and pulmonary function. Secondary objectives include exploring their relationships with airway inflammation and asthma control status. This study seeks to establish a non-invasive, objective, and quantifiable approach to asthma phenotyping, providing evidence for integrating traditional diagnostic features with modern clinical data to support precision management of pediatric asthma.

This study is a single-center, observational cohort study designed to investigate the clinical value of multimodal tongue and pulse characteristics in the phenotypic classification and disease monitoring of pediatric asthma. The study will be conducted at Shanghai Children's Medical Center and will enroll children aged 5-18 years with a confirmed diagnosis of asthma who are undergoing routine outpatient follow-up. Standardized tongue images and pulse wave data will be collected using validated tongue diagnosis and pulse diagnosis devices during outpatient visits in which pulmonary function testing is clinically indicated. Tongue image acquisition will be performed under controlled lighting and positioning conditions to ensure reproducibility, while pulse wave signals will be recorded using standardized sensor placement and acquisition protocols. These procedures are non-invasive and are integrated into routine clinical workflows without altering standard medical care. Quantitative features will be extracted from tongue images and pulse waveforms using image processing and signal analysis methods, including color space transformation, texture analysis, brightness distribution, and pulse waveform time-domain and frequency-domain parameters. These multimodal features will be systematically linked to clinical data, including asthma clinical stage (acute exacerbation, chronic persistent phase, or clinical remission), pulmonary function parameters, such as Forced Expiratory Volume in 1 second(FEV₁), Forced Expiratory Volume in 1 second/Forced Vital Capacity(FEV₁/FVC), Resistance at 5 Hz(R5), and Resistance at 20 Hz(R20), fractional exhaled nitric oxide when available, blood eosinophil counts, allergic sensitization status, and Asthma Control Questionnaire-5 (ACQ-5) scores. The primary objective of the study is to evaluate the associations between multimodal tongue-pulse quantitative features and asthma clinical stages and pulmonary function status. Secondary objectives include exploring the relationships between tongue-pulse features and airway inflammation, asthma control level, and allergic phenotype, as well as assessing their potential utility in phenotypic stratification and longitudinal disease monitoring. Participants will be followed according to routine clinical practice, with repeated data collection occurring only when pulmonary function testing is performed as part of standard care. Longitudinal analyses will be conducted using time-stamped clinical and tongue-pulse data to account for irregular visit intervals. This study aims to establish a non-invasive, objective, and quantifiable framework for integrating traditional diagnostic features with modern clinical indicators, providing evidence to support precision phenotyping and individualized management of pediatric asthma.