Brief Report: The ADOS Calibrated Severity Score Best Measures Autism Diagnostic Symptom Severity in Pre-School Children / Lisa D. WIGGINS in Journal of Autism and Developmental Disorders, 49-7 (July 2019)  

Public ISBD [article]  inJournal of Autism and Developmental Disorders > 49-7 (July 2019) . - p.2999-3006 Titre : Brief Report: The ADOS Calibrated Severity Score Best Measures Autism Diagnostic Symptom Severity in Pre-School Children Type de document : Texte imprimé et/ou numérique Auteurs : Lisa D. WIGGINS, Auteur ; Brian D. BARGER, Auteur ; E. MOODY, Auteur ; G. SOKE, Auteur ; J. PANDEY, Auteur ; S. LEVY, Auteur Article en page(s) : p.2999-3006 Langues : Anglais (eng) Mots-clés : Autism Diagnostic Observation Schedule  Autism spectrum disorder  Calibrated severity score  Symptom severity Index. décimale : PER Périodiques Résumé : The severity of autism spectrum disorder (ASD) is often measured by co-occurring conditions, such as intellectual disability or language delay, rather than deficits in social interaction, and restricted interests and repetitive behaviors. The Autism Diagnostic Observation Schedule calibrated severity score (ADOS CSS) was created to facilitate comparison of the diagnostic features of ASD independent of related conditions over time. We examined the relationship between the ADOS CSS, ADOS total score, and clinician rated degree of impairment (DOI) in the Study to Explore Early Development. Like others, we confirmed that, among the measures we evaluated, the ADOS CSS was least influenced by developmental functioning and demographic factors and is therefore the best measure of core features of ASD in pre-school children. En ligne : http://dx.doi.org/10.1007/s10803-017-3072-x Permalink : https://www.cra-rhone-alpes.org/cid/opac_css/index.php?lvl=notice_display&id=402 [article] Brief Report: The ADOS Calibrated Severity Score Best Measures Autism Diagnostic Symptom Severity in Pre-School Children [Texte imprimé et/ou numérique] / Lisa D. WIGGINS, Auteur ; Brian D. BARGER, Auteur ; E. MOODY, Auteur ; G. SOKE, Auteur ; J. PANDEY, Auteur ; S. LEVY, Auteur . - p.2999-3006. Langues : Anglais (eng) in Journal of Autism and Developmental Disorders > 49-7 (July 2019) . - p.2999-3006 Mots-clés : Autism Diagnostic Observation Schedule  Autism spectrum disorder  Calibrated severity score  Symptom severity Index. décimale : PER Périodiques Résumé : The severity of autism spectrum disorder (ASD) is often measured by co-occurring conditions, such as intellectual disability or language delay, rather than deficits in social interaction, and restricted interests and repetitive behaviors. The Autism Diagnostic Observation Schedule calibrated severity score (ADOS CSS) was created to facilitate comparison of the diagnostic features of ASD independent of related conditions over time. We examined the relationship between the ADOS CSS, ADOS total score, and clinician rated degree of impairment (DOI) in the Study to Explore Early Development. Like others, we confirmed that, among the measures we evaluated, the ADOS CSS was least influenced by developmental functioning and demographic factors and is therefore the best measure of core features of ASD in pre-school children. En ligne : http://dx.doi.org/10.1007/s10803-017-3072-x Permalink : https://www.cra-rhone-alpes.org/cid/opac_css/index.php?lvl=notice_display&id=402

Infection and Fever in Pregnancy and Autism Spectrum Disorders: Findings from the Study to Explore Early Development / Lisa A. CROEN in Autism Research, 12-10 (October 2019)  

Public ISBD [article]  inAutism Research > 12-10 (October 2019) . - p.1551-1561 Titre : Infection and Fever in Pregnancy and Autism Spectrum Disorders: Findings from the Study to Explore Early Development Type de document : Texte imprimé et/ou numérique Auteurs : Lisa A. CROEN, Auteur ; Y. QIAN, Auteur ; Paul ASHWOOD, Auteur ; O. ZERBO, Auteur ; Diana SCHENDEL, Auteur ; J. PINTO-MARTIN, Auteur ; M. DANIELE FALLIN, Auteur ; S. LEVY, Auteur ; Laura A. SCHIEVE, Auteur ; M. YEARGIN-ALLSOPP, Auteur ; Katherine R. SABOURIN, Auteur ; Jennifer L. AMES, Auteur Article en page(s) : p.1551-1561 Langues : Anglais (eng) Mots-clés : autism  developmental disorder  immune function  infection  neurodevelopment  prenatal Index. décimale : PER Périodiques Résumé : Maternal infection and fever during pregnancy have been implicated in the etiology of autism spectrum disorder (ASD); however, studies have not been able to separate the effects of fever itself from the impact of a specific infectious organism on the developing brain. We utilized data from the Study to Explore Early Development (SEED), a case-control study among 2- to 5-year-old children born between 2003 and 2006 in the United States, to explore a possible association between maternal infection and fever during pregnancy and risk of ASD and other developmental disorders (DDs). Three groups of children were included: children with ASD (N = 606) and children with DDs (N = 856), ascertained from clinical and educational sources, and children from the general population (N = 796), randomly sampled from state birth records. Information about infection and fever during pregnancy was obtained from a telephone interview with the mother shortly after study enrollment and maternal prenatal and labor/delivery medical records. ASD and DD status was determined by an in-person standardized developmental assessment of the child at 3-5 years of age. After adjustment for covariates, maternal infection anytime during pregnancy was not associated with ASD or DDs. However, second trimester infection accompanied by fever elevated risk for ASD approximately twofold (aOR = 2.19, 95% confidence interval 1.14-4.23). These findings of an association between maternal infection with fever in the second trimester and increased risk of ASD in the offspring suggest that the inflammatory response to the infectious agent may be etiologically relevant. Autism Res 2019, 12: 1551-1561. (c) 2019 International Society for Autism Research, Wiley Periodicals, Inc. LAY SUMMARY: Using data from a large multisite study in the United States-the Study to Explore Early Development-we found that women who had an infection during the second trimester of pregnancy accompanied by a fever are more likely to have children with ASD. These findings suggest the possibility that only more severe infections accompanied by a robust inflammatory response increase the risk of ASD. En ligne : http://dx.doi.org/10.1002/aur.2175 Permalink : https://www.cra-rhone-alpes.org/cid/opac_css/index.php?lvl=notice_display&id=408 [article] Infection and Fever in Pregnancy and Autism Spectrum Disorders: Findings from the Study to Explore Early Development [Texte imprimé et/ou numérique] / Lisa A. CROEN, Auteur ; Y. QIAN, Auteur ; Paul ASHWOOD, Auteur ; O. ZERBO, Auteur ; Diana SCHENDEL, Auteur ; J. PINTO-MARTIN, Auteur ; M. DANIELE FALLIN, Auteur ; S. LEVY, Auteur ; Laura A. SCHIEVE, Auteur ; M. YEARGIN-ALLSOPP, Auteur ; Katherine R. SABOURIN, Auteur ; Jennifer L. AMES, Auteur . - p.1551-1561. Langues : Anglais (eng) in Autism Research > 12-10 (October 2019) . - p.1551-1561 Mots-clés : autism  developmental disorder  immune function  infection  neurodevelopment  prenatal Index. décimale : PER Périodiques Résumé : Maternal infection and fever during pregnancy have been implicated in the etiology of autism spectrum disorder (ASD); however, studies have not been able to separate the effects of fever itself from the impact of a specific infectious organism on the developing brain. We utilized data from the Study to Explore Early Development (SEED), a case-control study among 2- to 5-year-old children born between 2003 and 2006 in the United States, to explore a possible association between maternal infection and fever during pregnancy and risk of ASD and other developmental disorders (DDs). Three groups of children were included: children with ASD (N = 606) and children with DDs (N = 856), ascertained from clinical and educational sources, and children from the general population (N = 796), randomly sampled from state birth records. Information about infection and fever during pregnancy was obtained from a telephone interview with the mother shortly after study enrollment and maternal prenatal and labor/delivery medical records. ASD and DD status was determined by an in-person standardized developmental assessment of the child at 3-5 years of age. After adjustment for covariates, maternal infection anytime during pregnancy was not associated with ASD or DDs. However, second trimester infection accompanied by fever elevated risk for ASD approximately twofold (aOR = 2.19, 95% confidence interval 1.14-4.23). These findings of an association between maternal infection with fever in the second trimester and increased risk of ASD in the offspring suggest that the inflammatory response to the infectious agent may be etiologically relevant. Autism Res 2019, 12: 1551-1561. (c) 2019 International Society for Autism Research, Wiley Periodicals, Inc. LAY SUMMARY: Using data from a large multisite study in the United States-the Study to Explore Early Development-we found that women who had an infection during the second trimester of pregnancy accompanied by a fever are more likely to have children with ASD. These findings suggest the possibility that only more severe infections accompanied by a robust inflammatory response increase the risk of ASD. En ligne : http://dx.doi.org/10.1002/aur.2175 Permalink : https://www.cra-rhone-alpes.org/cid/opac_css/index.php?lvl=notice_display&id=408

Sparsifying machine learning models identify stable subsets of predictive features for behavioral detection of autism / S. LEVY in Molecular Autism, 8 (2017)  

Public ISBD [article]  inMolecular Autism > 8 (2017) . - 65p. Titre : Sparsifying machine learning models identify stable subsets of predictive features for behavioral detection of autism Type de document : Texte imprimé et/ou numérique Auteurs : S. LEVY, Auteur ; M. DUDA, Auteur ; N. HABER, Auteur ; Dennis P. WALL, Auteur Article en page(s) : 65p. Langues : Anglais (eng) Mots-clés : Asd  Autism  Autism diagnosis  Autism screening  Autism spectrum disorder  Machine learning  Sparse machine learning  company focused on building digital solutions for child health.Springer Nature  remains neutral with regard to jurisdictional claims in published maps and  institutional affiliations. Index. décimale : PER Périodiques Résumé : Background: Autism spectrum disorder (ASD) diagnosis can be delayed due in part to the time required for administration of standard exams, such as the Autism Diagnostic Observation Schedule (ADOS). Shorter and potentially mobilized approaches would help to alleviate bottlenecks in the healthcare system. Previous work using machine learning suggested that a subset of the behaviors measured by ADOS can achieve clinically acceptable levels of accuracy. Here we expand on this initial work to build sparse models that have higher potential to generalize to the clinical population. Methods: We assembled a collection of score sheets for two ADOS modules, one for children with phrased speech (Module 2; 1319 ASD cases, 70 controls) and the other for children with verbal fluency (Module 3; 2870 ASD cases, 273 controls). We used sparsity/parsimony enforcing regularization techniques in a nested cross validation grid search to select features for 17 unique supervised learning models, encoding missing values as additional indicator features. We augmented our feature sets with gender and age to train minimal and interpretable classifiers capable of robust detection of ASD from non-ASD. Results: By applying 17 unique supervised learning methods across 5 classification families tuned for sparse use of features and to be within 1 standard error of the optimal model, we find reduced sets of 10 and 5 features used in a majority of models. We tested the performance of the most interpretable of these sparse models, including Logistic Regression with L2 regularization or Linear SVM with L1 regularization. We obtained an area under the ROC curve of 0.95 for ADOS Module 3 and 0.93 for ADOS Module 2 with less than or equal to 10 features. Conclusions: The resulting models provide improved stability over previous machine learning efforts to minimize the time complexity of autism detection due to regularization and a small parameter space. These robustness techniques yield classifiers that are sparse, interpretable and that have potential to generalize to alternative modes of autism screening, diagnosis and monitoring, possibly including analysis of short home videos. En ligne : http://dx.doi.org/10.1186/s13229-017-0180-6 Permalink : https://www.cra-rhone-alpes.org/cid/opac_css/index.php?lvl=notice_display&id=330 [article] Sparsifying machine learning models identify stable subsets of predictive features for behavioral detection of autism [Texte imprimé et/ou numérique] / S. LEVY, Auteur ; M. DUDA, Auteur ; N. HABER, Auteur ; Dennis P. WALL, Auteur . - 65p. Langues : Anglais (eng) in Molecular Autism > 8 (2017) . - 65p. Mots-clés : Asd  Autism  Autism diagnosis  Autism screening  Autism spectrum disorder  Machine learning  Sparse machine learning  company focused on building digital solutions for child health.Springer Nature  remains neutral with regard to jurisdictional claims in published maps and  institutional affiliations. Index. décimale : PER Périodiques Résumé : Background: Autism spectrum disorder (ASD) diagnosis can be delayed due in part to the time required for administration of standard exams, such as the Autism Diagnostic Observation Schedule (ADOS). Shorter and potentially mobilized approaches would help to alleviate bottlenecks in the healthcare system. Previous work using machine learning suggested that a subset of the behaviors measured by ADOS can achieve clinically acceptable levels of accuracy. Here we expand on this initial work to build sparse models that have higher potential to generalize to the clinical population. Methods: We assembled a collection of score sheets for two ADOS modules, one for children with phrased speech (Module 2; 1319 ASD cases, 70 controls) and the other for children with verbal fluency (Module 3; 2870 ASD cases, 273 controls). We used sparsity/parsimony enforcing regularization techniques in a nested cross validation grid search to select features for 17 unique supervised learning models, encoding missing values as additional indicator features. We augmented our feature sets with gender and age to train minimal and interpretable classifiers capable of robust detection of ASD from non-ASD. Results: By applying 17 unique supervised learning methods across 5 classification families tuned for sparse use of features and to be within 1 standard error of the optimal model, we find reduced sets of 10 and 5 features used in a majority of models. We tested the performance of the most interpretable of these sparse models, including Logistic Regression with L2 regularization or Linear SVM with L1 regularization. We obtained an area under the ROC curve of 0.95 for ADOS Module 3 and 0.93 for ADOS Module 2 with less than or equal to 10 features. Conclusions: The resulting models provide improved stability over previous machine learning efforts to minimize the time complexity of autism detection due to regularization and a small parameter space. These robustness techniques yield classifiers that are sparse, interpretable and that have potential to generalize to alternative modes of autism screening, diagnosis and monitoring, possibly including analysis of short home videos. En ligne : http://dx.doi.org/10.1186/s13229-017-0180-6 Permalink : https://www.cra-rhone-alpes.org/cid/opac_css/index.php?lvl=notice_display&id=330

---

1 (1 - 3 / 3)