Table 2.
Overview of the studies included.
| References | Study | Dataset | Study population size [%male] | Age range, age mean [SD] |
|---|---|---|---|---|
| Ball et al. (2017) | Modelling neuroanatomical variation during childhood and adolescence with neighbourhood-preserving embedding. | PING-Study | 768 [53%] | 3–21 y, 12.3 y |
| Ball et al. (2021) | Individual variation underlying brain age estimates in typical development | PING-Study | 768 [53%] | 3–21 y, 12.28 [5.02] y |
| Ball et al. (2019) | Charting shared developmental trajectories of cortical thickness and structural connectivity in childhood and adolescence | PING-Study | 456 [51.1%] | 3.2–21.0 y, 12.6 [4.91] y |
| Brown et al. (2017) | Prediction of brain network age and factors of delayed maturation in very preterm infants | self-recruited | 115 | 27 and 45 w PMA |
| Brown et al. (2012) | Neuroanatomical assessment of biological maturity | PING-Study | 885 [52.2%] | 3–20 y, 13.0 [4.9] y |
| Cao et al. (2015) | Development and validation of a brain maturation index using longitudinal neuroanatomical scans. | National Institute of Health (NIH) pediatric repository | 303 | 4.88–18.35 y |
| Chen et al. (2022) | Deep learning to predict neonatal and infant brain age from myelination on brain MRI scans. | self-recruited | 469 | 0–25 m GCA, 65.0 [32] w |
| Chen et al. (2022) | 438 | 0–25 m GCA, 64.4 [30] w | ||
| Chen et al. (2022) | 389 | 0–25 m GCA, 61.9 [29] | ||
| Chung et al. (2018) | Use of machine learning to determine deviance in neuroanatomical maturity associated with future psychosis in youths at clinically high risk. | PING-Study | 953 [51.7%] | 3–21 y |
| Erus et al. (2015) | Imaging patterns of brain development and their relationship to cognition | Philadelphia Neurodevelopmental Cohort | 621 [43.5%] | 8–22 y, 15.08 y [3.27] |
| Franke et al. (2012) | Brain maturation: predicting individual BrainAGE in children and adolescents using structural MRI | National Institute of Health (NIH) pediatric repository | 394 [52.5%] | 5–18 y |
| Galdi et al. (2020) | Neonatal morphometric similarity mapping for predicting brain age and characterizing neuroanatomic variation associated with preterm birth | Self-recruited | 105 [52.3%] | 38–45 w |
| Gschwandtner et al. (2020) | Deep learning for estimation of functional brain maturation from EEG of premature neonates. | Self-recruited | 43 | 24–42 w |
| He et al. (2020) | Brain age estimation using LSTM on children’s brain MRI | National Institute of Health (NIH) pediatric repository Massachusetts General and Boston Children’s Hospitals |
1,212 | 0–22 y |
| 428 | 0–6 y | |||
| Hosseinzadeh Kassani et al. (2020) | Causality-based feature fusion for brain neuro-developmental analysis. | Philadelphia Neurodevelopmental Cohort | 1,445 | 8–21 y |
| Hu et al. (2020) | Hierarchical rough-to-fine model for infant age prediction based on cortical features. | Self-recruited | 50 | 1, 3, 6, 9, 12, 18 and 24 m |
| Hu et al. (2021) | Accurate brain age prediction model for healthy children and adolescents using 3D-CNN and dimensional attention | ABIDE I ABIDE II ADHD200 |
880 | 6–18 y, 11.8y [2.8] |
| Kardan et al. (2022) | Resting-state functional connectivity identifies individuals and predicts age in 8- to 26-month-olds. | Baby Connectome Project | 112 [53.6%] | 8–26 m, 15.7 [5.2] |
| Kawahara et al. (2017) | BrainNetCNN: Convolutional neural networks for brain networks; towards predicting neurodevelopment | Self-recruited | 115 | 24 and 32 m PMA |
| Kelly et al. (2022) | Investigating brain structural maturation in children and adolescents born very preterm using the brain age framework. | VIBeS PING |
768 [52.6%] | 3–21 y, 12.28 y, |
| Khundrakpam et al. (2015) | Prediction of brain maturity based on cortical thickness at different spatial resolutions | National Institute of Health (NIH) pediatric repository | 308 [44.2%] | Range not given, 12.9 y [3.8] |
| Li et al. (2018) | Brain age prediction based on resting-state functional connectivity patterns using convolutional neural networks | Philadelphia Neurodevelopmental Cohort | 983 | 8–22 y |
| Lund et al. (2022) | Brain age prediction using fMRI network coupling in youths and associations with psychiatric symptoms. | Philadelphia Neurodevelopmental Cohort Healthy Brain Network |
1,126 | 8–22 y |
| Morita et al. (2022b) | Pediatric brain CT image segmentation methods for effective age prediction models | Self-recruited | 204 | 0–47 m |
| Morita et al. (2022a) | Quantification of pediatric brain development with X-ray CT images using 3D-CNN | Self-recruited | 204 | 0–47 m |
| Nielsen et al. (2019) | Evaluating the prediction of brain maturity from functional connectivity after motion artifact denoising. | Self-recruited | 122 [54.1%] | 7–31 y |
| O’Toole et al. (2016) | Estimating functional brain maturity in very and extremely preterm neonates using automated analysis of the electroencephalogram | Self-recruited | 49 | 23–32 w GCA |
| Qu et al. (2020) | BAENET: A brain age estimation network with 3D skipping and Outlier constraint loss | ABIDE II ADHD200 HBN |
1915 | 5–18 y |
| Saha et al. (2018) | Investigating brain age deviation in preterm infants: a deep learning approach | Self-recruited | 86 | 29–47 w PMA |
| Shabanian et al. (2019) | Classification of neurodevelopmental age in normal infants using 3D-CNN based on brain MRI | NIMH Data Archive | 112 | 8 d – 3 y |
| Smyser et al. (2016) | Prediction of brain maturity in infants using machine-learning algorithms. | Self-recruited | 50 | Preterm: 36–41 w PMA, 38 w [1 w] |
| 50 | Term: 37–41 w PMA, 39 w [1 w] |
|||
| Stevenson et al. (2017) | Functional maturation in preterm infants measured by serial recording of cortical activity | Self-recruited | 43 [41.9%] | 25–38 w PMA, 25.6 w |
| Stevenson et al. (2020) | Reliability and accuracy of EEG interpretation for estimating age in preterm infants. | Self-recruited | 62 | 25–38 weeks PMA |
| Vandenbosch et al. (2019) | EEG-based age-prediction models as stable and heritable indicators of brain maturational level in children and adolescents. | Self-recruited | Dataset 1: 836 | 5 and 7 y 16 and 18 y |
| Dataset 2: 621 | 12, 14 and 16 y | |||
| Zhao T. et al. (2019) | Unbiased age-specific structural brain atlases for Chinese pediatric population. | Peking University Dataset Beijing HuiLongGuan ADHD200 |
Dataset 1: 328 | 6–12 y, 9.03 [1.36] |
| Dataset 2: 114 | 6–12 y, 9.06 [1.38] | |||
| Dataset 3: 71 | 8–12 y, 10.26 [1.78] | |||
| Sturmfels et al. (2018) | A domain guided CNN architecture for predicting age from structural brain images | Philadelphia Neurodevelopmental Cohort | 724 | 8–21 y |
| Hong et al. (2020) | Brain age prediction of children using routine brain MR images via deep learning | Self-recruited | 220 | 0 to 5 y |
| Zhao Y. et al. (2019) | Brain age prediction: Cortical and subcortical shape covariation in the developing human brain | Healthy Brain Network Nathan Kline Institute - Rockland Sample |
869 (60.9%) | 5.02–17.95 y |
| 210 (58.1%) | 6.68–17.94 y | |||
| Liang et al. (2019) | Investigating systematic bias in brain age estimation with application to post-traumatic stress disorders | ABIDE CoRR DLBS |
566 | 6 to 89 y |
| 778 | ||||
| 315 | ||||
| Lewis et al. (2018) | T1 white/gray contrast as a predictor of chronological age, and an index of cognitive performance | NIH Pediatric data PING |
401 | 4.5–18.5 y |
| 760 | 3–20 y | |||
| Dean et al. (2015) | Estimating the age of healthy infants from quantitative myelin water fraction maps | Self-recruited | 209 (58.9%) | 76–1,526 d |
| Pardoe and Kuzniecky (2018) | NAPR: a cloud-based framework for neuroanatomical age prediction | ABIDE, ABIDE II, CoRR, DLBS, and NKI Rockland dataset | 2,367 | Not specified besides figure |
| Lavanga et al. (2018) | A brain-age model for preterm infants based on functional connectivity. | Self-recruited | 30 | 27–42 w |
| Liu et al. (2024) | Brain age predicted using graph convolutional neural network explains neurodevelopmental trajectory in preterm neonates | University of California at San Francisco) Benioff Children’s Hospital (UCSF) developing Human Connectome Project |
129 | 32.1–43.4 w |
| 407 (54.1%) | 29–45 w | |||
| Tang et al. (2023) | A deep learning-based brain age prediction model for preterm infants via neonatal MRI | Self-recruited | 281 | 27–37 w, 33.4 w |
| Liu et al. (2023) | Brain age prediction in children aged 0–5 years based on T1 magnetic resonance images | Self-recruited | 290 | 0–5 y |
| Mendes et al. (2023) | Generalizability of 3D CNN models for age estimation in diverse youth populations using structural MRI. | ABIDE-II ADHD-200 ABCD BHRCS |
580 | 6.1–20.0 |
| 922 | 7.1–19.9 | |||
| 11,031 | 5.8–14.3 | |||
| 737 | 8.9–11.1 | |||
| Zandvoort et al. (2024) | Sensory event-related potential morphology predicts age in premature infants. | research database John Radcliffe Hospital | 82 | 28–40 w PMA |
| Nielsen et al. (2023) | Maturation of large-scale brain systems over the first month of life. | eLABE CUDDEL O2P2 |
262 | At birth |
| 45 | 0–18 m | |||
| 5 | 0-72 h | |||
| Hu et al. (2023) | MRI-based brain age prediction model for children under 3 years old using deep residual network. | Self-recruited | 658 (62.9%), 230 diseased | 0–1,092 d |
| Griffiths-King et al. (2023) | Predicting ‘Brainage’ in late childhood to adolescence (6–17 yrs) using structural MRI, morphometric similarity, and machine learning. | Autism Brain Imaging Data Exchange cohort from the Pre-Procecessed Connectome Project | 327 (79.2%) | 6.5–16.9,12.4 [± 2.5] |
| Bellantuono et al. (2021) | Predicting brain age with complex networks: From adolescence to adulthood | ABIDE | 1,112 | 7 to 64 y |
Study population size is presented in total and the percentage of males, if given. Age range is further broken down to mean and standard deviation (SD), if presented by authors. Years (y), months (m), weeks (w), post menstrual age (PMA), gestation corrected age (GCA).