Brain maturation: predicting individual BrainAGE in children and adolescents using structural MRI

Area of Science:

• Neuroscience
• Developmental Biology
• Medical Imaging

Background:

• Human brain development involves complex, age-dependent changes in gray matter, white matter, and volume.
• A validated reference curve for predicting structural brain maturation in children and adolescents is currently lacking.
• Structural magnetic resonance imaging (MRI) data offers a means to study these developmental processes.

Purpose of the Study:

• To develop and validate a reference curve for structural brain maturation in children and adolescents.
• To establish a clinically valuable tool for predicting brain age based on MRI data.

Main Methods:

• Utilized kernel regression methods to create the BrainAGE framework.
• Applied the BrainAGE framework to a dataset of 394 healthy children and adolescents from the NIH MRI study of normal brain development.
• Validated the framework by correlating predicted brain age with chronological age.

Main Results:

• The BrainAGE framework accurately predicted individual brain maturation, with a correlation of r=0.93 between predicted and chronological age.
• The mean absolute error in age prediction was 1.1 years.
• Predicted brain age successfully differentiated between all developmental age groups and showed potential in preterm-born adolescents.

Conclusions:

• The novel BrainAGE approach provides a validated reference curve for structural brain maturation.
• This method has the potential to be clinically valuable for detecting normal and abnormal brain maturation.
• The BrainAGE framework offers important prognostic information for brain development.