Infant Brain Age Estimation With T1w/T2w Ratio MRI: A Myelination-Aware Deep Learning Approach.

Authors

Abstract

Brain age estimation provides a noninvasive MRI biomarker of neurodevelopment. In infancy, rapid regionally ordered myelination reflects brain maturation, yet early-life brain age estimation remains underexplored, particularly with myelination-sensitive MRI and biologically informed modeling. To develop and evaluate a biologically informed deep learning framework for infant brain age estimation using T1w/T2w ratio MRI. Retrospective. Internal cohort: 629 infants aged 0-24 months (626 with age-appropriate myelination, train/validation/test = 376/125/125), 3 with myelin-related developmental abnormalities for qualitative review. External cohort: 10 healthy infants aged 0-15 months (5 females, 5 males). Internal: 3T; 3D gradient-echo or 2D spin-echo T1w, and 2D turbo spin-echo T2w. External: 3T; 3D gradient-echo T1w and 2D turbo spin-echo T2w. 3D convolutional neural networks were trained with T1w, T2w, and T1w/T2w ratio inputs using manually defined biological age labels from visual myelination assessment. The model incorporated multi-task learning for age regression, white matter segmentation, and image reconstruction. Performance was evaluated using five-fold cross-validation with repeated random splits. Metrics included mean absolute error, root mean squared error, <math xmlns="http://www.w3.org/1998/Math/MathML"> <semantics> <mrow><msup><mi>R</mi> <mn>2</mn></msup> </mrow> <annotation>$$ {R}^2 $$</annotation></semantics> </math> , and Pearson and Spearman correlations. Modality differences were tested using one-way ANOVA, <math xmlns="http://www.w3.org/1998/Math/MathML"> <semantics><mrow><mi>t</mi></mrow> <annotation>$$ t $$</annotation></semantics> </math> -tests, and Mann-Whitney <math xmlns="http://www.w3.org/1998/Math/MathML"> <semantics><mrow><mi>U</mi></mrow> <annotation>$$ U $$</annotation></semantics> </math> , with Cohen's <math xmlns="http://www.w3.org/1998/Math/MathML"> <semantics><mrow><mi>d</mi></mrow> <annotation>$$ d $$</annotation></semantics> </math> and 95% confidence intervals. In the external cohort, absolute prediction errors were compared using the Wilcoxon signed-rank test. Statistical significance was defined as <math xmlns="http://www.w3.org/1998/Math/MathML"> <semantics><mrow><mi>p</mi> <mo><</mo> <mn>0.05</mn></mrow> <annotation>$$ p<0.05 $$</annotation></semantics> </math> . T1w/T2w ratio models achieved the best overall performance (MAE: 1.489  <math xmlns="http://www.w3.org/1998/Math/MathML"> <semantics><mrow><mo>±</mo></mrow> <annotation>$$ \pm $$</annotation></semantics> </math> 0.302 months; <math xmlns="http://www.w3.org/1998/Math/MathML"> <semantics><mrow><mi>r</mi></mrow> <annotation>$$ r $$</annotation></semantics> </math>  = 0.966  <math xmlns="http://www.w3.org/1998/Math/MathML"> <semantics><mrow><mo>±</mo></mrow> <annotation>$$ \pm $$</annotation></semantics> </math>  0.012), compared with T1w (2.055  <math xmlns="http://www.w3.org/1998/Math/MathML"> <semantics><mrow><mo>±</mo></mrow> <annotation>$$ \pm $$</annotation></semantics> </math>  0.944; 0.933  <math xmlns="http://www.w3.org/1998/Math/MathML"> <semantics><mrow><mo>±</mo></mrow> <annotation>$$ \pm $$</annotation></semantics> </math>  0.061), T2w (1.794  <math xmlns="http://www.w3.org/1998/Math/MathML"> <semantics><mrow><mo>±</mo></mrow> <annotation>$$ \pm $$</annotation></semantics> </math>  0.434; 0.947  <math xmlns="http://www.w3.org/1998/Math/MathML"> <semantics><mrow><mo>±</mo></mrow> <annotation>$$ \pm $$</annotation></semantics> </math>  0.023), T1w+T2w (1.546  <math xmlns="http://www.w3.org/1998/Math/MathML"> <semantics><mrow><mo>±</mo></mrow> <annotation>$$ \pm $$</annotation></semantics> </math>  0.291; 0.960  <math xmlns="http://www.w3.org/1998/Math/MathML"> <semantics><mrow><mo>±</mo></mrow> <annotation>$$ \pm $$</annotation></semantics> </math>  0.013), and T1w+T2w+RI (1.498  <math xmlns="http://www.w3.org/1998/Math/MathML"> <semantics><mrow><mo>±</mo></mrow> <annotation>$$ \pm $$</annotation></semantics> </math>  0.313; 0.963 <math xmlns="http://www.w3.org/1998/Math/MathML"> <semantics><mrow><mo>±</mo></mrow> <annotation>$$ \pm $$</annotation></semantics> </math> 0.012). Modality effects were significant for MAE, RMSE, <math xmlns="http://www.w3.org/1998/Math/MathML"> <semantics> <mrow><msup><mi>R</mi> <mn>2</mn></msup> </mrow> <annotation>$$ {R}^2 $$</annotation></semantics> </math> , <math xmlns="http://www.w3.org/1998/Math/MathML"> <semantics><mrow><mi>r</mi></mrow> <annotation>$$ r $$</annotation></semantics> </math> , but not for <math xmlns="http://www.w3.org/1998/Math/MathML"> <semantics><mrow><mi>ρ</mi></mrow> <annotation>$$ \rho $$</annotation></semantics> </math> ( <math xmlns="http://www.w3.org/1998/Math/MathML"> <semantics><mrow><mi>p</mi> <mo>=</mo> <mn>0.250</mn></mrow> <annotation>$$ p=0.250 $$</annotation></semantics> </math> ). Auxiliary-task and multi-scale modeling numerically improved performance (MAE, 1.203 months; <math xmlns="http://www.w3.org/1998/Math/MathML"> <semantics><mrow><mi>r</mi></mrow> <annotation>$$ r $$</annotation></semantics> </math>  = 0.979). External validation showed the lowest error for the RI-based model (MAE, 1.16 months), and Grad-CAM highlighted myelination-relevant white matter. T1w/T2w ratio MRI combined with biologically informed deep learning enabled accurate and interpretable infant brain age estimation. This framework showed promising cross-scanner performance and may support MRI-based assessment of early brain maturation. 3. 2.

Tags

Topics