Related Experiment Videos
Machine-Learning Models Outperform Clinicians in Predicting Postnatal Growth Failure Among Very Low Birth Weight
Joohee Lim1, Sook Hyun Park2, Teahyen Cha1
1Department of Pediatrics, Yonsei University College of Medicine, Seoul 03722, Republic of Korea.
Insights
A machine-learning model accurately predicted postnatal growth failure (PGF) in preterm infants, outperforming neonatologists. This AI tool can aid in early risk stratification and targeted nutritional support for very low birth weight (VLBW) infants.
Area of Science:
- Neonatalogy
- Artificial Intelligence in Medicine
- Biostatistics
Background:
- Postnatal growth failure (PGF) in preterm infants is linked to poor neurodevelopmental outcomes.
- Predicting PGF is challenging due to multiple influencing clinical factors.
- Machine learning (ML) offers potential for predicting complex neonatal outcomes.
Purpose of the Study:
- To compare the predictive performance of neonatologists with an ML model for PGF detection.
- To evaluate the efficacy of an extreme gradient boosting (XGBoost) model in identifying PGF risk.
Main Methods:
- Defined PGF as a weight z-score decrease >1.28 from birth to discharge.
- Trained an XGBoost ML model on data from 7954 very low birth weight (VLBW) infants.
- Assessed 100 clinical cases (50 with PGF) using nine neonatologists and the XGBoost model, evaluating seven performance metrics.
Main Results:
- The XGBoost model achieved a higher prediction score (79/100) than neonatologists (median 52/100).
- XGBoost demonstrated superior performance with an AUROC of 0.79 and F1 score of 0.80.
- The ML model showed a significantly lower error rate (0.21) compared to neonatologists (0.49).
Conclusions:
- ML models exhibit superior or comparable predictive performance to neonatologists for PGF detection.
- ML-based prediction tools can enhance early risk stratification for VLBW infants.
- These models support timely and targeted nutritional interventions in preterm neonates.
Abstract:
Background/Objectives: Early detection of postnatal growth failure (PGF) is essential for optimizing nutritional management in preterm infants, as PGF is associated with adverse neurodevelopmental outcomes. Early prediction remains difficult because postnatal growth is influenced by multiple clinical factors including gestation age, birth weight, nutritional status, and comorbidities. Machine-learning approaches have been proposed to predict complex neonatal outcomes. This study compared the predictive performance of neonatologists with that of a machine-learning model for predicting PGF. Methods: PGF was defined as a decrease in weight z-score greater than 1.28 at discharge compared with birth. A machine-learning model based on extreme gradient boosting (XGBoost) was trained using a dataset of 7954 very low birth weight (VLBW) infants. Nine neonatologists independently assessed 100 clinical cases through a questionnaire-based evaluation, including 50 patients with PGF. Predictive performance was evaluated using seven metrics: area under the receiver operating characteristic curve (AUROC), accuracy, error rate, positive predictive value (PPV), sensitivity, specificity, and F1 score. Results: The neonatologists had a median of 5 years (range: 4-10 years) of clinical experience. The median prediction score among the neonatologists was 52/100 (range, 44-60), whereas the XGBoost model achieved 79/100. The XGBoost model achieved an AUROC of 0.79, accuracy of 0.79, error rate of 0.21, sensitivity of 0.82, and an F1 score of 0.80, demonstrating superior overall performance compared to the neonatologists. In addition, the XGBoost model had a lower error rate than the neonatologists (0.21 vs. 0.49), whereas specificity (0.76 vs. 0.86) and PPV (0.77 vs. 0.53) did not differ significantly. Conclusions: The machine-learning model demonstrated superior or comparable predictive performance to that of neonatologists in detecting PGF. Machine-learning-based prediction models may support early risk stratification and targeted nutritional management in VLBW infants.
