Prediction of Autism at 3 Years from Behavioural and Developmental Measures in High-Risk Infants: A Longitudinal Cross-Domain Classifier Analysis

G. Bussu; E. J.H. Jones; T. Charman; M. H. Johnson; J. K. Buitelaar; The Basis Team; S. Baron-Cohen; R. Bedford; P. Bolton; A. Blasi; S. Chandler; C. Cheung; K. Davies; M. Elsabbagh; J. Fernandes; I. Gammer; H. Garwood; T. Gliga; J. Guiraud; K. Hudry; M. H. Johnson; M. Liew; S. Lloyd-Fox; H. Maris; L. O’Hara; G. Pasco; A. Pickles; H. Ribeiro; E. Salomone; L. Tucker; A. Volein

doi:10.1007/s10803-018-3509-x

Prediction of Autism at 3 Years from Behavioural and Developmental Measures in High-Risk Infants: A Longitudinal Cross-Domain Classifier Analysis

G. Bussu^*, E. J.H. Jones, T. Charman, M. H. Johnson, J. K. Buitelaar, The Basis Team, S. Baron-Cohen, R. Bedford, P. Bolton, A. Blasi, S. Chandler, C. Cheung, K. Davies, M. Elsabbagh, J. Fernandes, I. Gammer, H. Garwood, T. Gliga, J. Guiraud, K. HudryM. H. Johnson, M. Liew, S. Lloyd-Fox, H. Maris, L. O’Hara, G. Pasco, A. Pickles, H. Ribeiro, E. Salomone, L. Tucker, A. Volein

^*Corresponding author for this work

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

58 Citations (Scopus)

205 Downloads (Pure)

Abstract

We integrated multiple behavioural and developmental measures from multiple time-points using machine learning to improve early prediction of individual Autism Spectrum Disorder (ASD) outcome. We examined Mullen Scales of Early Learning, Vineland Adaptive Behavior Scales, and early ASD symptoms between 8 and 36 months in high-risk siblings (HR; n = 161) and low-risk controls (LR; n = 71). Longitudinally, LR and HR-Typical showed higher developmental level and functioning, and fewer ASD symptoms than HR-Atypical and HR-ASD. At 8 months, machine learning classified HR-ASD at chance level, and broader atypical development with 69.2% Area Under the Curve (AUC). At 14 months, ASD and broader atypical development were classified with approximately 71% AUC. Thus, prediction of ASD was only possible with moderate accuracy at 14 months.

Original language	English
Pages (from-to)	1-16
Number of pages	16
Journal	Journal of Autism and Developmental Disorders
Early online date	16 Feb 2018
DOIs	https://doi.org/10.1007/s10803-018-3509-x
Publication status	E-pub ahead of print - 16 Feb 2018

Keywords

Autism
Data integration
Early prediction
High-risk
Individual prediction
Longitudinal study
Machine learning

Access to Document

10.1007/s10803-018-3509-xLicence: CC BY

Prediction of Autism at 3_BUSSU_Publishedonline16February2018_GOLD VoR (CC BY)Final published version, 1.71 MBLicence: CC BY

Cite this

Bussu, G., Jones, E. J. H., Charman, T., Johnson, M. H., Buitelaar, J. K., The Basis Team, Baron-Cohen, S., Bedford, R., Bolton, P., Blasi, A., Chandler, S., Cheung, C., Davies, K., Elsabbagh, M., Fernandes, J., Gammer, I., Garwood, H., Gliga, T., Guiraud, J., ... Volein, A. (2018). Prediction of Autism at 3 Years from Behavioural and Developmental Measures in High-Risk Infants: A Longitudinal Cross-Domain Classifier Analysis. Journal of Autism and Developmental Disorders, 1-16. Advance online publication. https://doi.org/10.1007/s10803-018-3509-x

@article{07225c284e424ee1ac7f12e20036eb33,

title = "Prediction of Autism at 3 Years from Behavioural and Developmental Measures in High-Risk Infants: A Longitudinal Cross-Domain Classifier Analysis",

abstract = "We integrated multiple behavioural and developmental measures from multiple time-points using machine learning to improve early prediction of individual Autism Spectrum Disorder (ASD) outcome. We examined Mullen Scales of Early Learning, Vineland Adaptive Behavior Scales, and early ASD symptoms between 8 and 36 months in high-risk siblings (HR; n = 161) and low-risk controls (LR; n = 71). Longitudinally, LR and HR-Typical showed higher developmental level and functioning, and fewer ASD symptoms than HR-Atypical and HR-ASD. At 8 months, machine learning classified HR-ASD at chance level, and broader atypical development with 69.2% Area Under the Curve (AUC). At 14 months, ASD and broader atypical development were classified with approximately 71% AUC. Thus, prediction of ASD was only possible with moderate accuracy at 14 months.",

keywords = "Autism, Data integration, Early prediction, High-risk, Individual prediction, Longitudinal study, Machine learning",

author = "G. Bussu and Jones, {E. J.H.} and T. Charman and Johnson, {M. H.} and Buitelaar, {J. K.} and {The Basis Team} and S. Baron-Cohen and R. Bedford and P. Bolton and A. Blasi and S. Chandler and C. Cheung and K. Davies and M. Elsabbagh and J. Fernandes and I. Gammer and H. Garwood and T. Gliga and J. Guiraud and K. Hudry and Johnson, {M. H.} and M. Liew and S. Lloyd-Fox and H. Maris and L. O{\textquoteright}Hara and G. Pasco and A. Pickles and H. Ribeiro and E. Salomone and L. Tucker and A. Volein",

year = "2018",

month = feb,

day = "16",

doi = "10.1007/s10803-018-3509-x",

language = "English",

pages = "1--16",

journal = "Journal of Autism and Developmental Disorders",

issn = "0162-3257",

publisher = "Springer",

}

Bussu, G, Jones, EJH, Charman, T, Johnson, MH, Buitelaar, JK, The Basis Team, Baron-Cohen, S, Bedford, R , Bolton, P , Blasi, A , Chandler, S , Cheung, C, Davies, K, Elsabbagh, M, Fernandes, J, Gammer, I, Garwood, H, Gliga, T, Guiraud, J, Hudry, K, Johnson, MH, Liew, M, Lloyd-Fox, S, Maris, H, O’Hara, L, Pasco, G , Pickles, A, Ribeiro, H, Salomone, E, Tucker, L & Volein, A 2018, 'Prediction of Autism at 3 Years from Behavioural and Developmental Measures in High-Risk Infants: A Longitudinal Cross-Domain Classifier Analysis', Journal of Autism and Developmental Disorders, pp. 1-16. https://doi.org/10.1007/s10803-018-3509-x

TY - JOUR

T1 - Prediction of Autism at 3 Years from Behavioural and Developmental Measures in High-Risk Infants

T2 - A Longitudinal Cross-Domain Classifier Analysis

AU - Bussu, G.

AU - Jones, E. J.H.

AU - Charman, T.

AU - Johnson, M. H.

AU - Buitelaar, J. K.

AU - The Basis Team

AU - Baron-Cohen, S.

AU - Bedford, R.

AU - Bolton, P.

AU - Blasi, A.

AU - Chandler, S.

AU - Cheung, C.

AU - Davies, K.

AU - Elsabbagh, M.

AU - Fernandes, J.

AU - Gammer, I.

AU - Garwood, H.

AU - Gliga, T.

AU - Guiraud, J.

AU - Hudry, K.

AU - Johnson, M. H.

AU - Liew, M.

AU - Lloyd-Fox, S.

AU - Maris, H.

AU - O’Hara, L.

AU - Pasco, G.

AU - Pickles, A.

AU - Ribeiro, H.

AU - Salomone, E.

AU - Tucker, L.

AU - Volein, A.

PY - 2018/2/16

Y1 - 2018/2/16

N2 - We integrated multiple behavioural and developmental measures from multiple time-points using machine learning to improve early prediction of individual Autism Spectrum Disorder (ASD) outcome. We examined Mullen Scales of Early Learning, Vineland Adaptive Behavior Scales, and early ASD symptoms between 8 and 36 months in high-risk siblings (HR; n = 161) and low-risk controls (LR; n = 71). Longitudinally, LR and HR-Typical showed higher developmental level and functioning, and fewer ASD symptoms than HR-Atypical and HR-ASD. At 8 months, machine learning classified HR-ASD at chance level, and broader atypical development with 69.2% Area Under the Curve (AUC). At 14 months, ASD and broader atypical development were classified with approximately 71% AUC. Thus, prediction of ASD was only possible with moderate accuracy at 14 months.

AB - We integrated multiple behavioural and developmental measures from multiple time-points using machine learning to improve early prediction of individual Autism Spectrum Disorder (ASD) outcome. We examined Mullen Scales of Early Learning, Vineland Adaptive Behavior Scales, and early ASD symptoms between 8 and 36 months in high-risk siblings (HR; n = 161) and low-risk controls (LR; n = 71). Longitudinally, LR and HR-Typical showed higher developmental level and functioning, and fewer ASD symptoms than HR-Atypical and HR-ASD. At 8 months, machine learning classified HR-ASD at chance level, and broader atypical development with 69.2% Area Under the Curve (AUC). At 14 months, ASD and broader atypical development were classified with approximately 71% AUC. Thus, prediction of ASD was only possible with moderate accuracy at 14 months.

KW - Autism

KW - Data integration

KW - Early prediction

KW - High-risk

KW - Individual prediction

KW - Longitudinal study

KW - Machine learning

UR - http://www.scopus.com/inward/record.url?scp=85042108327&partnerID=8YFLogxK

U2 - 10.1007/s10803-018-3509-x

DO - 10.1007/s10803-018-3509-x

M3 - Article

AN - SCOPUS:85042108327

SN - 0162-3257

SP - 1

EP - 16

JO - Journal of Autism and Developmental Disorders

JF - Journal of Autism and Developmental Disorders

ER -

Prediction of Autism at 3 Years from Behavioural and Developmental Measures in High-Risk Infants: A Longitudinal Cross-Domain Classifier Analysis

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this