Abstract:Background Assessment of appropriate brain myelination on T1- and T2-weighted MRI scans is based on gestationally corrected age (GCA) and requires subjective visual inspection of the brain with knowledge of normal myelination milestones. Purpose To develop a convolutional neural network (CNN) capable of estimating neonatal and infant GCA based on brain myelination on MRI scans. Materials and methods In this retrospective study from one academic medical center, brain MRI scans of patients aged 0-25 months with reported normal myelination were consecutively collected between January 1995 and June 2019. The GCA at MRI was manually calculated. After exclusion criteria were applied, T1- and T2-weighted MRI scans were preprocessed with skull stripping, linear registration, z scoring for normalization, and downsampling. A three-dimensional regression CNN was trained to predict GCA using mean absolute error (MAE) as its loss function. Attention maps were calculated using layer-wise relevance propagation. Models were validated on an external test set from the National Institutes of Health (NIH). Model MAEs were compared using Kruskal-Wallis and Mann-Whitney tests. Results A total of 518 neonates and infants (mean GCA, 67 weeks ± 33 [SD], 56% male) was included, comprising 469 T1-, 438 T2-, and 389 T1- and T2-weighted studies. Across 10 runs, MAEs of T1-, T2-, and T1- and T2-weighted networks were 9.8 ± 2.3, 9.1 ± 1.9, and 7.7 ± 1.7 weeks, respectively. Attention map analysis demonstrated increased network attention to the cerebellum, posterior white matter, and basal ganglia signal in neonates with GCA of less than 40 weeks and the anterior white matter signal in infants with GCA of more than 120 weeks, corresponding to the known progression of myelination. The T1- and T2-weighted network tested on the external NIH test set had an MAE of 9.1 weeks, which was reduced to 5.9 weeks with further training using half the external test set (P < .001). Conclusion A three-dimensional convolutional neural network can predict the gestationally corrected age of neonates and infants aged 0-25 months based on brain myelination patterns on T1- and T2-weighted MRI scans. © RSNA, 2022 Online supplemental material is available for this article.

Multi-view Attention for gestational age at birth prediction

Automated 3D Fetal Brain Segmentation Using an Optimized Deep Learning Approach

Multi-modal Perceptual Adversarial Learning for Longitudinal Prediction of Infant MR Images

Deep Learning with Attention to Predict Gestational Age of the Fetal Brain

Attention-guided deep learning for gestational age prediction using fetal brain MRI

Prediction of fetal brain gestational age using multihead attention with Xception

PDFF‐CNN: An attention‐guided dynamic multi‐orientation feature fusion method for gestational age prediction on imbalanced fetal brain MRI dataset

Geometric Deep Learning for Post-Menstrual Age Prediction based on the Neonatal White Matter Cortical Surface

MRI Volume-Based Robust Brain Age Estimation Using Weight-Shared Spatial Attention in 3D CNNs

Predicting gestational age at birth in the context of preterm birth from multi-modal fetal MRI

Attention-modulated Multi-Branch Convolutional Neural Networks for Neonatal Brain Tissue Segmentation

Automatic cortical surface parcellation in the fetal brain using attention-gated spherical U-net

GestAltNet: aggregation and attention to improve deep learning of gestational age from placental whole-slide images

Multi-scale Multi-hierarchy Attention Convolutional Neural Network for Fetal Brain Extraction

Multi-Task Learning Approach for Unified Biometric Estimation from Fetal Ultrasound Anomaly Scans

Unsupervised Segmentation of Fetal Brain MRI using Deep Learning Cascaded Registration

Prediction of Infant Cognitive Development with Cortical Surface-Based Multimodal Learning

Multi-task Learning for Neonatal Brain Segmentation Using 3D Dense-Unet with Dense Attention Guided by Geodesic Distance

Fetal MRI by robust deep generative prior reconstruction and diffeomorphic registration: application to gestational age prediction

Deep CNN ensembles and suggestive annotations for infant brain MRI segmentation

Deep Learning to Predict Neonatal and Infant Brain Age from Myelination on Brain MRI Scans