Peer-reviewed article
High angular resolution diffusion imaging of neurodevelopment in children through data creation with deep learning.
Oladosu O, Lo F, Geeraert B, Lebel C, Zhang Y ·
Summary
Written by AI, not yet AICS's view · claude-haiku-4-5-20251001This is a technical neuroscience paper about using deep learning to create high-resolution brain imaging data from lower-resolution scans in children aged 2-10 years. The study demonstrates that artificial intelligence can predict detailed brain scans from simpler data, potentially reducing scan time. The research shows that predicted imaging data produce similar patterns to actual scans and reveal age-related and sex-specific brain development patterns. This is methodological research on imaging technology rather than an evaluation of an intervention for child well-being.
Key points
- Deep learning successfully predicted high-resolution brain diffusion imaging from lower-resolution source data
- Model performance improved with larger training samples, larger window sizes, and additional brain segmentation inputs
- Predicted and source-only imaging data showed similar patterns across white matter tracts and metrics
- Brain tract volumes increased with age and showed consistent sex differences
- The approach could potentially reduce pediatric brain scan time by half
Limitations
This is a technical feasibility study focused on imaging methodology, not an evaluation of health or developmental outcomes for children. It does not measure the impact on child well-being, health status, or development beyond describing brain structure patterns.
Abstract
High-angular resolution diffusion imaging (HARDI) is an advanced method for characterizing brain microstructure and function. However, HARDI is time-consuming limiting real-world applications, especially in children. We aimed to address the challenge by creating non-acquired HARDI data through deep learning and testing utility in neurodevelopment. Brain diffusion magnetic resonance imaging from 95 children aged 2-10 years (49 females) were examined. Each subject included two source b-value datasets: 750 s/mm2 and 2000 s/mm2, 30 directions each with five b0 volumes. Deep learning was voxel-wise, approximately 1,620,000 voxels from 12-subject each matched on age and sex for training and validation, respectively, to predict b2000 s/mm2 from b750 s/mm2 dataset. Different training samples, window sizes, and tissue inputs were also assessed. The remaining 71 individuals were used for tract-based analysis of neurodevelopment, which involved eight brain white matter tracts. Results showed that model performance improved with an increasing training sample, large window size, and additional brain segmentation input. HARDI outcomes including predicted data were similar in pattern to source-only data across tracts and metrics, with all appeared to increase with age. Additionally, tract volumes showed consistent sex differences between prediction-integrated and source-only datasets in all but pyramidal tract. Collectively, deep learning-enabled HARDI is feasible for pediatric imaging, which may help reduce scan time by half and characterize sex-specific neurodevelopment.