Age estimates from brain magnetic resonance images of children younger than two years of age using deep learning

• Published in: Magnetic resonance imaging Vol. 79; pp. 38 - 44
• Main Authors: Kawaguchi, Masahiro, Kidokoro, Hiroyuki, Ito, Rintaro, Shiraki, Anna, Suzuki, Takeshi, Maki, Yuki, Tanaka, Masaharu, Sakaguchi, Yoko, Yamamoto, Hiroyuki, Takahashi, Yosiyuki, Naganawa, Shinji, Natsume, Jun
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier Inc 01.06.2021
• Subjects: Age estimation; Brain MRI; Convolutional neural network; Deep learning; Infant; Deep learning; Infant; Age estimation; Brain MRI; Convolutional neural network
• ISSN: 0730-725X; 1873-5894; 1873-5894
• DOI: 10.1016/j.mri.2021.03.004

The accuracy of brain age estimates from magnetic resonance (MR) images has improved with the advent of deep learning artificial intelligence (AI) models. However, most previous studies on predicting age emphasized aging from childhood to adulthood and old age, and few studies have focused on early brain development in children younger than 2 years of age. Here, we performed brain age estimates based on MR images in children younger than 2 years of age using deep learning. Our AI model, developed with one slice each of raw T1- and T2-weighted images from each subject, estimated brain age with a mean absolute error of 8.2 weeks (1.9 months). The estimates of our AI model were close to those of human specialists. The AI model also estimated the brain age of subjects with a myelination delay as significantly younger than the chronological age. These results indicate that the prediction accuracy of our AI model approached that of human specialists and that our simple method requiring less data and preprocessing facilitates a radiological assessment of brain development, such as monitoring maturational changes in myelination. •A deep learning AI model estimated brain age from magnetic resonance images from early infancy.•With minimal information and preprocessing, the AI model predicted age with similar accuracy to human specialists.•The AI model estimated age from brain images with delayed myelination as younger than the chronological age.•The AI model facilitates radiological assessment in the clinical setting.