Development and validation of a machine learning-based tool to predict autism among children / Kim Steven BETTS in Autism Research, 16-5 (May 2023)  

Public ISBD [article]  inAutism Research > 16-5 (May 2023) . - p.941-952 Titre : Development and validation of a machine learning-based tool to predict autism among children Type de document : Texte imprimé et/ou numérique Auteurs : Kim Steven BETTS, Auteur ; Kevin CHAI, Auteur ; Steve KISELY, Auteur ; Rosa ALATI, Auteur Article en page(s) : p.941-952 Langues : Anglais (eng) Index. décimale : PER Périodiques Résumé : Abstract Autism is a lifelong condition for which intervention must occur as early as possible to improve social functioning. Thus, there is great interest in improving our ability to diagnose autism as early as possible. We take a novel approach to this challenge by combining machine learning with maternal and infant health administrative data to construct a prediction model capable of predicting autism disorder (defined as ICD10 84.0) in the general population. The sample included all mother-offspring pairs from the Australian state of New South Wales (NSW) between January 2003 and December 2005 (n = 262,650 offspring), linked across three health administrative data sets including the NSW perinatal data collection (PDC); the NSW admitted patient data collection (APDC) and the NSW mental health ambulatory data collection (MHADC). Our most successful model was able to predict autism disorder with an area under the receiver operating curve of 0.73, with the strongest risk factors for diagnoses found to include offspring gender, maternal age at birth, delivery analgesia, maternal prenatal tobacco disorders, and low 5-min APGAR score. Our findings indicate that the combination of machine learning and routinely collected admin data, with further refinement and increased accuracy than achieved by us, may play a role in the early detection of autism disorders. En ligne : http://dx.doi.org/https://doi.org/10.1002/aur.2912 Permalink : https://www.cra-rhone-alpes.org/cid/opac_css/index.php?lvl=notice_display&id=503 [article] Development and validation of a machine learning-based tool to predict autism among children [Texte imprimé et/ou numérique] / Kim Steven BETTS, Auteur ; Kevin CHAI, Auteur ; Steve KISELY, Auteur ; Rosa ALATI, Auteur . - p.941-952. Langues : Anglais (eng) in Autism Research > 16-5 (May 2023) . - p.941-952 Index. décimale : PER Périodiques Résumé : Abstract Autism is a lifelong condition for which intervention must occur as early as possible to improve social functioning. Thus, there is great interest in improving our ability to diagnose autism as early as possible. We take a novel approach to this challenge by combining machine learning with maternal and infant health administrative data to construct a prediction model capable of predicting autism disorder (defined as ICD10 84.0) in the general population. The sample included all mother-offspring pairs from the Australian state of New South Wales (NSW) between January 2003 and December 2005 (n = 262,650 offspring), linked across three health administrative data sets including the NSW perinatal data collection (PDC); the NSW admitted patient data collection (APDC) and the NSW mental health ambulatory data collection (MHADC). Our most successful model was able to predict autism disorder with an area under the receiver operating curve of 0.73, with the strongest risk factors for diagnoses found to include offspring gender, maternal age at birth, delivery analgesia, maternal prenatal tobacco disorders, and low 5-min APGAR score. Our findings indicate that the combination of machine learning and routinely collected admin data, with further refinement and increased accuracy than achieved by us, may play a role in the early detection of autism disorders. En ligne : http://dx.doi.org/https://doi.org/10.1002/aur.2912 Permalink : https://www.cra-rhone-alpes.org/cid/opac_css/index.php?lvl=notice_display&id=503

Exploring the relationships between pathogen-specific prenatal infections requiring inpatient admission and domains of offspring behaviour at age 5 / Steve KISELY ; Rosa ALATI in Journal of Child Psychology and Psychiatry, 65-9 (September 2024)  

Public ISBD [article]  inJournal of Child Psychology and Psychiatry > 65-9 (September 2024) . - p.1213-1222 Titre : Exploring the relationships between pathogen-specific prenatal infections requiring inpatient admission and domains of offspring behaviour at age 5 Type de document : Texte imprimé et/ou numérique Auteurs : Steve KISELY, Auteur ; Rosa ALATI, Auteur Article en page(s) : p.1213-1222 Langues : Anglais (eng) Index. décimale : PER Périodiques Résumé : Background Research exploring the relationship between prenatal infection and child behavioural outcomes would benefit from further studies utilising full-population samples with the scale to investigate specific infections and to employ robust designs. We tested the association among several common infections requiring inpatient admission during and after pregnancy with a range of childhood behavioural outcomes, to determine whether any negative impact was specific to the period of foetal development. Methods The sample included all mother-offspring pairs from the Australian state of New South Wales (NSW) for whom the child commenced their first year of full-time schooling in 2009 (~age 5?years; n = 77,302 offspring), with records linked across four health administrative data sets including the NSW perinatal data collection (PDC), the NSW admitted patient data collection (APDC) and the NSW component of the 2009 Australian Early Development Census (AEDC). Multivariable linear regression was used to test associations between a number of infections requiring inpatient admission during and after pregnancy with a range of teacher assessed behavioural outcomes. Results Associations specific to the prenatal period were only found for streptococcus A although this would need to be reproduced in external samples given the low prevalence. Otherwise, 12 out of 15 selected infections either showed no association prenatally or also demonstrated associations in the 12?months after pregnancy. For example, prenatal hepatitis C, influenza and urinary E. coli infections were associated with lower scores of several domains of childhood behaviour, but even stronger associations were found when these same maternal infections occurred after pregnancy. Conclusions The prenatal infections we tested appeared not to impact childhood behaviour by altering foetal neurodevelopment. Rather, the strong associations we found among infections occurring during and after pregnancy point to either residual socioeconomic/lifestyle factors or a shared familial/genetic liability between infections and behavioural problems. En ligne : https://doi.org/10.1111/jcpp.13964 Permalink : https://www.cra-rhone-alpes.org/cid/opac_css/index.php?lvl=notice_display&id=534 [article] Exploring the relationships between pathogen-specific prenatal infections requiring inpatient admission and domains of offspring behaviour at age 5 [Texte imprimé et/ou numérique] / Steve KISELY, Auteur ; Rosa ALATI, Auteur . - p.1213-1222. Langues : Anglais (eng) in Journal of Child Psychology and Psychiatry > 65-9 (September 2024) . - p.1213-1222 Index. décimale : PER Périodiques Résumé : Background Research exploring the relationship between prenatal infection and child behavioural outcomes would benefit from further studies utilising full-population samples with the scale to investigate specific infections and to employ robust designs. We tested the association among several common infections requiring inpatient admission during and after pregnancy with a range of childhood behavioural outcomes, to determine whether any negative impact was specific to the period of foetal development. Methods The sample included all mother-offspring pairs from the Australian state of New South Wales (NSW) for whom the child commenced their first year of full-time schooling in 2009 (~age 5?years; n = 77,302 offspring), with records linked across four health administrative data sets including the NSW perinatal data collection (PDC), the NSW admitted patient data collection (APDC) and the NSW component of the 2009 Australian Early Development Census (AEDC). Multivariable linear regression was used to test associations between a number of infections requiring inpatient admission during and after pregnancy with a range of teacher assessed behavioural outcomes. Results Associations specific to the prenatal period were only found for streptococcus A although this would need to be reproduced in external samples given the low prevalence. Otherwise, 12 out of 15 selected infections either showed no association prenatally or also demonstrated associations in the 12?months after pregnancy. For example, prenatal hepatitis C, influenza and urinary E. coli infections were associated with lower scores of several domains of childhood behaviour, but even stronger associations were found when these same maternal infections occurred after pregnancy. Conclusions The prenatal infections we tested appeared not to impact childhood behaviour by altering foetal neurodevelopment. Rather, the strong associations we found among infections occurring during and after pregnancy point to either residual socioeconomic/lifestyle factors or a shared familial/genetic liability between infections and behavioural problems. En ligne : https://doi.org/10.1111/jcpp.13964 Permalink : https://www.cra-rhone-alpes.org/cid/opac_css/index.php?lvl=notice_display&id=534

---

1 (1 - 2 / 2)