NEC-NET : Segmentation and Feature Extraction Network for the Neurocranium in Early Childhood

Area of Science:

• Pediatric neuroimaging
• Developmental biology
• Artificial intelligence in medicine

Background:

• The neurocranium rapidly develops in early life to house the growing brain.
• Disruptions in neurocranial maturation can stem from developmental issues or environmental factors like sleep position.
• Understanding typical neurocranial growth is crucial for identifying early signs of anomalies.

Purpose of the Study:

• To develop and validate a deep neural network pipeline, NEC-NET, for analyzing normative neurocranial development.
• To establish a baseline for early detection of neurodevelopmental abnormalities in children.
• To investigate age-based regional differences in infant neurocranial growth.

Main Methods:

• A deep neural network pipeline (NEC-NET) was designed, incorporating segmentation and classification modules.
• The pipeline was applied to T1-weighted Magnetic Resonance Imaging (MRI) data from healthy children aged 12 to 60 months.
• The method focused on optimizing neurocranium segmentation and analyzing normative growth patterns.

Main Results:

• The NEC-NET pipeline demonstrated optimized segmentation of the neurocranium.
• Preliminary results revealed distinct age-based regional differences in neurocranial development among infants.
• The study provides a foundation for analyzing typical neurocranial growth trajectories.

Conclusions:

• NEC-NET offers a novel approach for quantitative analysis of pediatric neurocranial development.
• The findings highlight the importance of age-specific normative data for identifying deviations.
• This AI-driven method has the potential to improve early detection of neurodevelopmental disorders.