Related Experiment Video
Updated: Jul 17, 2026
Measuring the Functional Abilities of Children Aged 3-6 Years Old with Observational Methods and Computer Tools
Published on: June 20, 2020
Development, internal and external evaluation of an artificial intelligence algorithm for child growth monitoring in
Pauline Scherdel1, Raphaële Houlbracq1, Emmanuel Lecoeur2,3
1Université Paris Cité and Université Sorbonne Paris Nord, Inserm, INRAE, Centre for Research in Epidemiology and StatisticS (CRESS), Paris, France.
Insights
An AI algorithm was developed to detect abnormal growth in children, potentially reducing diagnosis time for conditions like growth hormone deficiency (GHD) and Turner syndrome (TS). This AI tool shows high diagnostic performance, aiding early detection and improving pediatric growth monitoring.
Area of Science:
- Pediatric Endocrinology
- Medical Informatics
- Artificial Intelligence in Healthcare
Background:
- Accurate growth monitoring is crucial for identifying pediatric endocrine disorders.
- Early detection of conditions like growth hormone deficiency (GHD) and Turner syndrome (TS) can significantly impact treatment outcomes.
- Existing methods for growth monitoring may have limitations in early and accurate detection of abnormal growth trajectories.
Purpose of the Study:
- To develop and evaluate an artificial intelligence (AI) algorithm for detecting abnormal growth patterns in children.
- To assess the algorithm's diagnostic performance in identifying potential cases of GHD and TS.
- To determine the potential reduction in time to diagnosis for these conditions using the AI algorithm.
Main Methods:
- Utilized pre-diagnosis height measurements from children with GHD, TS, and healthy referents.
- Applied non-linear mixed models to create individual height growth curves.
- Employed multinomial logistic regression with age-specific models to predict abnormal growth trajectories.
- Validated the AI algorithm through internal 5-fold cross-validation and external evaluations.
Main Results:
- Developed five age-specific predictive models with high discrimination (AUROC 0.87-0.99) and good calibration.
- External evaluation demonstrated a cumulative sensitivity of 84.6% and specificity of 94.3%.
- The AI algorithm showed a median theoretical reduction in time to diagnosis of 2.0 years (1.6 years for GHD, 3.0 years for TS).
Conclusions:
- The developed AI algorithm exhibits high diagnostic performance for the early detection of GHD and TS in children.
- The AI tool has the potential to significantly shorten the time to diagnosis, enabling earlier intervention.
- Further refinement and broader external validation are recommended before widespread clinical implementation.
Abstract:
Our goal was to improve growth monitoring in children by developing and evaluating an artificial intelligence (AI) algorithm that can detect abnormal growth. We used pre-diagnosis height measurements for children with a diagnosis of growth hormone deficiency (GHD, n = 86) or Turner syndrome (TS, n = 87) in France (1990-2014) and all height measurements of apparently healthy children (referents, n = 923). We modeled the individual height growth curves by applying non-linear mixed models for each new measurement of each child. The resulting growth parameters were used in multinomial logistic regression across five pre-defined age ranges from 1 to 12 years to predict abnormal growth trajectories. For the five age-specific predictive models, we studied the discrimination and calibration curves, and retained the risk thresholds that offered a pre-defined specificity of 98%. Using all the available height measurements for cases and referents, we evaluated the cumulative diagnostic performance of the algorithm for detecting GHD or TS and the theoretical reduction in time to diagnosis. We evaluated these models internally using 5-fold cross-validation and externally from a regional sample of children with GHD (n = 77) or TS (n = 40) and a national sample of apparently healthy children (n = 5,755). The five age-specific predictive models had good discrimination (high AUROC range 0.87-0.99) and good calibration. Internal evaluations showed stable results. External evaluation revealed a cumulative sensitivity and specificity of 84.6% (95% CI 76.8-90.6) and 94.3% (93.6-94.9). The median theoretical reduction in time to diagnosis was 2.0 years (interquartile range 0.6-3.8): 1.6 years (0.5-2.8) for GHD and 3.0 years (1.0-5.4) for TS. To conclude, we developed and internally and externally evaluated an AI algorithm with high diagnostic performance for the early detection of GHD and TS. A refinement of the algorithm to include other target conditions and further external evaluation in other countries is needed before implementation in daily practice.
