Predicting age and clinical risk from the neonatal connectome

Yassine Taoudi-Benchekroun; Daan Christiaens; Irina Grigorescu; Oliver Gale-Grant; Andreas Schuh; Maximilian Pietsch; Andrew Chew; Nicholas  Harper; Shona Falconer; Tanya Poppe; Emer Hughes; Jana Hutter; Anthony Price; Jacques-Donald Tournier; Lucilio Cordero-Grande; Serena Counsell; Daniel Rueckert; Tomoki Arichi; Jo Hajnal; David Edwards; Maria Deprez; Dafnis Batalle

doi:https://doi.org/10.1016/j.neuroimage.2022.119319

Predicting age and clinical risk from the neonatal connectome

Yassine Taoudi-Benchekroun, Daan Christiaens, Irina Grigorescu, Oliver Gale-Grant, Andreas Schuh, Maximilian Pietsch, Andrew Chew, Nicholas Harper, Shona Falconer, Tanya Poppe, Emer Hughes, Jana Hutter, Anthony Price, Jacques-Donald Tournier, Lucilio Cordero-Grande, Serena Counsell, Daniel Rueckert, Tomoki Arichi, Jo Hajnal, David EdwardsMaria Deprez, Dafnis Batalle^*

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

13 Citations (Scopus)

186 Downloads (Pure)

Abstract

The development of perinatal brain connectivity underpins motor, cognitive and behavioural abilities in later life. Diffusion MRI allows the characterisation of subtle inter-individual differences in structural brain connectivity, though individual brain connectivity maps (connectomes) are by nature high in dimensionality and complex to interpret. Machine learning methods are a powerful tool to uncover properties of the connectome which are not readily visible and can give us clues as to how and why individual developmental trajectories differ. In this manuscript we used Deep Neural Networks and Random Forests to predict demographic and neurodevelopmental characteristics from neonatal structural connectomes in a large sample of babies (n = 524) from the developing Human Connectome Project. We achieved an accurate prediction of postmenstrual age (PMA) at scan in term-born infants (mean absolute error (MAE) = 0.72 weeks, r = 0.83 and p < 0.001). We also achieved good accuracy when predicting gestational age at birth in a cohort of term and preterm babies scanned at term equivalent age (MAE = 2.21 weeks, r = 0.82, p < 0.001). We subsequently used sensitivity analysis to obtain feature relevance from our prediction models, with the most important connections for prediction of PMA and GA found to predominantly involve frontal and temporal regions, thalami, and basal ganglia. From our models of PMA at scan for infants born at term, we computed a brain maturation index (predicted age minus actual age) of individual preterm neonates and found a significant correlation between this index and motor outcome at 18 months corrected age. Our results demonstrate the applicability of machine learning techniques in analyses of the neonatal connectome and suggest that a neural substrate of brain maturation with implications for future neurodevelopment is detectable at term equivalent age from the neonatal connectome.

Original language	English
Article number	119319
Journal	NeuroImage
Volume	257
DOIs	https://doi.org/10.1016/j.neuroimage.2022.119319
Publication status	Published - 15 Aug 2022

Access to Document

https://doi.org/10.1016/j.neuroimage.2022.119319Licence: CC BY

Taoudi-BenchekrounEtAl_Neuroimage2022_PostprintAccepted author manuscript, 4.34 MBLicence: CC BY
2022_03_Taoudi-Benchekroun_NeuroimageFinal published version, 4.29 MBLicence: CC BY

Cite this

Taoudi-Benchekroun, Y., Christiaens, D., Grigorescu, I., Gale-Grant, O., Schuh, A., Pietsch, M., Chew, A., Harper, N., Falconer, S., Poppe, T., Hughes, E., Hutter, J., Price, A., Tournier, J.-D., Cordero-Grande, L., Counsell, S., Rueckert, D., Arichi, T., Hajnal, J., ... Batalle, D. (2022). Predicting age and clinical risk from the neonatal connectome. NeuroImage, 257, Article 119319. https://doi.org/10.1016/j.neuroimage.2022.119319

@article{21dc2d4bc7a64e8ea2cae1a121ec6d90,

title = "Predicting age and clinical risk from the neonatal connectome",

abstract = "The development of perinatal brain connectivity underpins motor, cognitive and behavioural abilities in later life. Diffusion MRI allows the characterisation of subtle inter-individual differences in structural brain connectivity, though individual brain connectivity maps (connectomes) are by nature high in dimensionality and complex to interpret. Machine learning methods are a powerful tool to uncover properties of the connectome which are not readily visible and can give us clues as to how and why individual developmental trajectories differ. In this manuscript we used Deep Neural Networks and Random Forests to predict demographic and neurodevelopmental characteristics from neonatal structural connectomes in a large sample of babies (n = 524) from the developing Human Connectome Project. We achieved an accurate prediction of postmenstrual age (PMA) at scan in term-born infants (mean absolute error (MAE) = 0.72 weeks, r = 0.83 and p < 0.001). We also achieved good accuracy when predicting gestational age at birth in a cohort of term and preterm babies scanned at term equivalent age (MAE = 2.21 weeks, r = 0.82, p < 0.001). We subsequently used sensitivity analysis to obtain feature relevance from our prediction models, with the most important connections for prediction of PMA and GA found to predominantly involve frontal and temporal regions, thalami, and basal ganglia. From our models of PMA at scan for infants born at term, we computed a brain maturation index (predicted age minus actual age) of individual preterm neonates and found a significant correlation between this index and motor outcome at 18 months corrected age. Our results demonstrate the applicability of machine learning techniques in analyses of the neonatal connectome and suggest that a neural substrate of brain maturation with implications for future neurodevelopment is detectable at term equivalent age from the neonatal connectome.",

author = "Yassine Taoudi-Benchekroun and Daan Christiaens and Irina Grigorescu and Oliver Gale-Grant and Andreas Schuh and Maximilian Pietsch and Andrew Chew and Nicholas Harper and Shona Falconer and Tanya Poppe and Emer Hughes and Jana Hutter and Anthony Price and Jacques-Donald Tournier and Lucilio Cordero-Grande and Serena Counsell and Daniel Rueckert and Tomoki Arichi and Jo Hajnal and David Edwards and Maria Deprez and Dafnis Batalle",

note = "Funding Information: This work was supported by the European Research Council under the European Union Seventh Framework Programme (FP/2007-2013)/ERC Grant Agreement no. 319456. The authors acknowledge infrastructure support from the National Institute for Health Research (NIHR) Mental Health Biomedical Research Centre (BRC) at South London, Maudsley NHS Foundation Trust, King's College London and the NIHR-BRC at Guys and St Thomas{\textquoteright} Hospitals NHS Foundation Trust (GSTFT). The authors also acknowledge support in part from the Wellcome Engineering and Physical Sciences Research Council (EPSRC) Centre for Medical Engineering at King's College London [WT 203148/Z/16/Z] and the Medical Research Council (UK) [MR/K006355/1 and MR/L011530/1]. Additional sources of support included the Sackler Institute for Translational Neurodevelopment at King's College London, the European Autism Interventions (EU-AIMS) trial and the EU AIMS-2-TRIALS, a European Innovative Medicines Initiative Joint Undertaking under Grant Agreements No. 115300 and 777394, the resources of which are composed of financial contributions from the European Union's Seventh Framework Programme (Grant FP7/2007–2013). DC is supported by a Flemish Research Foundation (FWO) Fellowship [12ZV420N]. OGG is supported by a grant from the UK Medical Research Council / Sackler Foundation [MR/P502108/1]. MD is supported by the Academy of Medical Sciences Springboard Award [SBF004\1040]. TA is supported by a MRC Clinician Scientist Fellowship [MR/P008712/1] and translation support award [MR/V036874/1]. ADE, TA, JOM received support from the Medical Research Council Centre for Neurodevelopmental Disorders, King's College London [MR/N026063/1]. DB received support from a Wellcome Trust Seed Award in Science [217316/Z/19/Z]. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health. The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. Publisher Copyright: {\textcopyright} 2022",

year = "2022",

month = aug,

day = "15",

doi = "https://doi.org/10.1016/j.neuroimage.2022.119319",

language = "English",

volume = "257",

journal = "NeuroImage",

issn = "1053-8119",

publisher = "ACADEMIC PRESS INC ELSEVIER SCIENCE",

}

Taoudi-Benchekroun, Y, Christiaens, D , Grigorescu, I , Gale-Grant, O, Schuh, A, Pietsch, M , Chew, A, Harper, N, Falconer, S , Poppe, T, Hughes, E, Hutter, J , Price, A , Tournier, J-D , Cordero-Grande, L , Counsell, S, Rueckert, D, Arichi, T , Hajnal, J , Edwards, D , Deprez, M & Batalle, D 2022, 'Predicting age and clinical risk from the neonatal connectome', NeuroImage, vol. 257, 119319. https://doi.org/10.1016/j.neuroimage.2022.119319

TY - JOUR

T1 - Predicting age and clinical risk from the neonatal connectome

AU - Taoudi-Benchekroun, Yassine

AU - Christiaens, Daan

AU - Grigorescu, Irina

AU - Gale-Grant, Oliver

AU - Schuh, Andreas

AU - Pietsch, Maximilian

AU - Chew, Andrew

AU - Harper, Nicholas

AU - Falconer, Shona

AU - Poppe, Tanya

AU - Hughes, Emer

AU - Hutter, Jana

AU - Price, Anthony

AU - Tournier, Jacques-Donald

AU - Cordero-Grande, Lucilio

AU - Counsell, Serena

AU - Rueckert, Daniel

AU - Arichi, Tomoki

AU - Hajnal, Jo

AU - Edwards, David

AU - Deprez, Maria

AU - Batalle, Dafnis

N1 - Funding Information: This work was supported by the European Research Council under the European Union Seventh Framework Programme (FP/2007-2013)/ERC Grant Agreement no. 319456. The authors acknowledge infrastructure support from the National Institute for Health Research (NIHR) Mental Health Biomedical Research Centre (BRC) at South London, Maudsley NHS Foundation Trust, King's College London and the NIHR-BRC at Guys and St Thomas’ Hospitals NHS Foundation Trust (GSTFT). The authors also acknowledge support in part from the Wellcome Engineering and Physical Sciences Research Council (EPSRC) Centre for Medical Engineering at King's College London [WT 203148/Z/16/Z] and the Medical Research Council (UK) [MR/K006355/1 and MR/L011530/1]. Additional sources of support included the Sackler Institute for Translational Neurodevelopment at King's College London, the European Autism Interventions (EU-AIMS) trial and the EU AIMS-2-TRIALS, a European Innovative Medicines Initiative Joint Undertaking under Grant Agreements No. 115300 and 777394, the resources of which are composed of financial contributions from the European Union's Seventh Framework Programme (Grant FP7/2007–2013). DC is supported by a Flemish Research Foundation (FWO) Fellowship [12ZV420N]. OGG is supported by a grant from the UK Medical Research Council / Sackler Foundation [MR/P502108/1]. MD is supported by the Academy of Medical Sciences Springboard Award [SBF004\1040]. TA is supported by a MRC Clinician Scientist Fellowship [MR/P008712/1] and translation support award [MR/V036874/1]. ADE, TA, JOM received support from the Medical Research Council Centre for Neurodevelopmental Disorders, King's College London [MR/N026063/1]. DB received support from a Wellcome Trust Seed Award in Science [217316/Z/19/Z]. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health. The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. Publisher Copyright: © 2022

PY - 2022/8/15

Y1 - 2022/8/15

N2 - The development of perinatal brain connectivity underpins motor, cognitive and behavioural abilities in later life. Diffusion MRI allows the characterisation of subtle inter-individual differences in structural brain connectivity, though individual brain connectivity maps (connectomes) are by nature high in dimensionality and complex to interpret. Machine learning methods are a powerful tool to uncover properties of the connectome which are not readily visible and can give us clues as to how and why individual developmental trajectories differ. In this manuscript we used Deep Neural Networks and Random Forests to predict demographic and neurodevelopmental characteristics from neonatal structural connectomes in a large sample of babies (n = 524) from the developing Human Connectome Project. We achieved an accurate prediction of postmenstrual age (PMA) at scan in term-born infants (mean absolute error (MAE) = 0.72 weeks, r = 0.83 and p < 0.001). We also achieved good accuracy when predicting gestational age at birth in a cohort of term and preterm babies scanned at term equivalent age (MAE = 2.21 weeks, r = 0.82, p < 0.001). We subsequently used sensitivity analysis to obtain feature relevance from our prediction models, with the most important connections for prediction of PMA and GA found to predominantly involve frontal and temporal regions, thalami, and basal ganglia. From our models of PMA at scan for infants born at term, we computed a brain maturation index (predicted age minus actual age) of individual preterm neonates and found a significant correlation between this index and motor outcome at 18 months corrected age. Our results demonstrate the applicability of machine learning techniques in analyses of the neonatal connectome and suggest that a neural substrate of brain maturation with implications for future neurodevelopment is detectable at term equivalent age from the neonatal connectome.

AB - The development of perinatal brain connectivity underpins motor, cognitive and behavioural abilities in later life. Diffusion MRI allows the characterisation of subtle inter-individual differences in structural brain connectivity, though individual brain connectivity maps (connectomes) are by nature high in dimensionality and complex to interpret. Machine learning methods are a powerful tool to uncover properties of the connectome which are not readily visible and can give us clues as to how and why individual developmental trajectories differ. In this manuscript we used Deep Neural Networks and Random Forests to predict demographic and neurodevelopmental characteristics from neonatal structural connectomes in a large sample of babies (n = 524) from the developing Human Connectome Project. We achieved an accurate prediction of postmenstrual age (PMA) at scan in term-born infants (mean absolute error (MAE) = 0.72 weeks, r = 0.83 and p < 0.001). We also achieved good accuracy when predicting gestational age at birth in a cohort of term and preterm babies scanned at term equivalent age (MAE = 2.21 weeks, r = 0.82, p < 0.001). We subsequently used sensitivity analysis to obtain feature relevance from our prediction models, with the most important connections for prediction of PMA and GA found to predominantly involve frontal and temporal regions, thalami, and basal ganglia. From our models of PMA at scan for infants born at term, we computed a brain maturation index (predicted age minus actual age) of individual preterm neonates and found a significant correlation between this index and motor outcome at 18 months corrected age. Our results demonstrate the applicability of machine learning techniques in analyses of the neonatal connectome and suggest that a neural substrate of brain maturation with implications for future neurodevelopment is detectable at term equivalent age from the neonatal connectome.

UR - http://www.scopus.com/inward/record.url?scp=85131125707&partnerID=8YFLogxK

U2 - https://doi.org/10.1016/j.neuroimage.2022.119319

DO - https://doi.org/10.1016/j.neuroimage.2022.119319

M3 - Article

SN - 1053-8119

VL - 257

JO - NeuroImage

JF - NeuroImage

M1 - 119319

ER -

Predicting age and clinical risk from the neonatal connectome

Abstract

Access to Document

Other files and links

Fingerprint

Cite this