Centre d'Information et de documentation du CRA Rhône-Alpes
CRA
Informations pratiques
-
Adresse
Centre d'information et de documentation
du CRA Rhône-Alpes
Centre Hospitalier le Vinatier
bât 211
95, Bd Pinel
69678 Bron CedexHoraires
Lundi au Vendredi
9h00-12h00 13h30-16h00Contact
Tél: +33(0)4 37 91 54 65
Mail
Fax: +33(0)4 37 91 54 37
-
Résultat de la recherche
2 recherche sur le mot-clé 'Natural language processing'
Affiner la recherche Générer le flux rss de la recherche
Partager le résultat de cette recherche Faire une suggestion
Combining voice and language features improves automated autism detection / Heather MACFARLANE in Autism Research, 15-7 (July 2022)
[article]
Titre : Combining voice and language features improves automated autism detection Type de document : Texte imprimé et/ou numérique Auteurs : Heather MACFARLANE, Auteur ; Alexandra C. SALEM, Auteur ; Liu CHEN, Auteur ; Meysam ASGARI, Auteur ; Eric FOMBONNE, Auteur Article en page(s) : p.1288-1300 Langues : Anglais (eng) Mots-clés : autism automated measures communication disfluency natural language processing pragmatic language prosody voice Index. décimale : PER Périodiques Résumé : Variability in expressive and receptive language, difficulty with pragmatic language, and prosodic difficulties are all features of autism spectrum disorder (ASD). Quantifying language and voice characteristics is an important step for measuring outcomes for autistic people, yet clinical measurement is cumbersome and costly. Using natural language processing (NLP) methods and a harmonic model of speech, we analyzed language transcripts and audio recordings to automatically classify individuals as ASD or non-ASD. One-hundred fifty-eight participants (88 ASD, 70 non-ASD) ages 7 to 17 were evaluated with the autism diagnostic observation schedule (ADOS-2), module 3. The ADOS-2 was transcribed following modified SALT guidelines. Seven automated language measures (ALMs) and 10 automated voice measures (AVMs) for each participant were generated from the transcripts and audio of one ADOS-2 task. The measures were analyzed using support vector machine (SVM; a binary classifier) and receiver operating characteristic (ROC). The AVM model resulted in an ROC area under the curve (AUC) of 0.7800, the ALM model an AUC of 0.8748, and the combined model a significantly improved AUC of 0.9205. The ALM model better detected ASD participants who were younger and had lower language skills and shorter activity time. ASD participants detected by the AVM model had better language profiles than those detected by the language model. In combination, automated measurement of language and voice characteristics successfully differentiated children with and without autism. This methodology could help design robust outcome measures for future research. LAY SUMMARY: People with autism often struggle with communication differences which traditional clinical measures and language tests cannot fully capture. Using language transcripts and audio recordings from 158 children ages 7 to 17, we showed that automated, objective language and voice measurements successfully predict the child's diagnosis. This methodology could help design improved outcome measures for research. En ligne : http://dx.doi.org/10.1002/aur.2733 Permalink : https://www.cra-rhone-alpes.org/cid/opac_css/index.php?lvl=notice_display&id=477
in Autism Research > 15-7 (July 2022) . - p.1288-1300[article] Combining voice and language features improves automated autism detection [Texte imprimé et/ou numérique] / Heather MACFARLANE, Auteur ; Alexandra C. SALEM, Auteur ; Liu CHEN, Auteur ; Meysam ASGARI, Auteur ; Eric FOMBONNE, Auteur . - p.1288-1300.
Langues : Anglais (eng)
in Autism Research > 15-7 (July 2022) . - p.1288-1300
Mots-clés : autism automated measures communication disfluency natural language processing pragmatic language prosody voice Index. décimale : PER Périodiques Résumé : Variability in expressive and receptive language, difficulty with pragmatic language, and prosodic difficulties are all features of autism spectrum disorder (ASD). Quantifying language and voice characteristics is an important step for measuring outcomes for autistic people, yet clinical measurement is cumbersome and costly. Using natural language processing (NLP) methods and a harmonic model of speech, we analyzed language transcripts and audio recordings to automatically classify individuals as ASD or non-ASD. One-hundred fifty-eight participants (88 ASD, 70 non-ASD) ages 7 to 17 were evaluated with the autism diagnostic observation schedule (ADOS-2), module 3. The ADOS-2 was transcribed following modified SALT guidelines. Seven automated language measures (ALMs) and 10 automated voice measures (AVMs) for each participant were generated from the transcripts and audio of one ADOS-2 task. The measures were analyzed using support vector machine (SVM; a binary classifier) and receiver operating characteristic (ROC). The AVM model resulted in an ROC area under the curve (AUC) of 0.7800, the ALM model an AUC of 0.8748, and the combined model a significantly improved AUC of 0.9205. The ALM model better detected ASD participants who were younger and had lower language skills and shorter activity time. ASD participants detected by the AVM model had better language profiles than those detected by the language model. In combination, automated measurement of language and voice characteristics successfully differentiated children with and without autism. This methodology could help design robust outcome measures for future research. LAY SUMMARY: People with autism often struggle with communication differences which traditional clinical measures and language tests cannot fully capture. Using language transcripts and audio recordings from 158 children ages 7 to 17, we showed that automated, objective language and voice measurements successfully predict the child's diagnosis. This methodology could help design improved outcome measures for research. En ligne : http://dx.doi.org/10.1002/aur.2733 Permalink : https://www.cra-rhone-alpes.org/cid/opac_css/index.php?lvl=notice_display&id=477 Linguistic markers of autism in girls: evidence of a "blended phenotype" during storytelling / J. BOORSE in Molecular Autism, 10 (2019)
[article]
Titre : Linguistic markers of autism in girls: evidence of a "blended phenotype" during storytelling Type de document : Texte imprimé et/ou numérique Auteurs : J. BOORSE, Auteur ; M. COLA, Auteur ; S. PLATE, Auteur ; L. YANKOWITZ, Auteur ; J. PANDEY, Auteur ; Robert T. SCHULTZ, Auteur ; Julia PARISH-MORRIS, Auteur Article en page(s) : 14 p. Langues : Anglais (eng) Mots-clés : Autism spectrum disorder Mentalizing Narratives Natural language processing Sex differences Social cognition Storytelling Word choice Index. décimale : PER Périodiques Résumé : Background: Narrative abilities are linked to social impairment in autism spectrum disorder (ASD), such that reductions in words about cognitive processes (e.g., think, know) are thought to reflect underlying deficits in social cognition, including Theory of Mind. However, research suggests that typically developing (TD) boys and girls tell narratives in sex-specific ways, including differential reliance on cognitive process words. Given that most studies of narration in ASD have been conducted in predominantly male samples, it is possible that prior results showing reduced cognitive processing language in ASD may not generalize to autistic girls. To answer this question, we measured the relative frequency of two kinds of words in stories told by autistic girls and boys: nouns (words that indicate object-oriented storytelling) and cognitive process words (words like think and know that indicate mentalizing or attention to other peoples' internal states). Methods: One hundred two verbally fluent school-aged children [girls with ASD (N = 21) and TD (N = 19), and boys with ASD (N = 41) and TD (N = 21)] were matched on age, IQ, and maternal education. Children told a story from a sequence of pictures, and word frequencies (nouns, cognitive process words) were compared. Results: Autistic children of both sexes consistently produced a greater number of nouns than TD controls, indicating object-focused storytelling. There were no sex differences in cognitive process word use in the TD group, but autistic girls produced significantly more cognitive process words than autistic boys, despite comparable autism symptom severity. Thus, autistic girls showed a unique narrative profile that overlapped with autistic boys and typical girls/boys. Noun use correlated significantly with parent reports of social symptom severity in all groups, but cognitive process word use correlated with social ability in boys only. Conclusion: This study extends prior research on autistic children's storytelling by measuring sex differences in the narratives of a relatively large, well-matched sample of children with and without ASD. Importantly, prior research showing that autistic children use fewer cognitive process words is true for boys only, while object-focused language is a sex-neutral linguistic marker of ASD. These findings suggest that sex-sensitive screening and diagnostic methods-preferably using objective metrics like natural language processing-may be helpful for identifying autistic girls, and could guide the development of future personalized treatment strategies. En ligne : http://dx.doi.org/10.1186/s13229-019-0268-2 Permalink : https://www.cra-rhone-alpes.org/cid/opac_css/index.php?lvl=notice_display&id=398
in Molecular Autism > 10 (2019) . - 14 p.[article] Linguistic markers of autism in girls: evidence of a "blended phenotype" during storytelling [Texte imprimé et/ou numérique] / J. BOORSE, Auteur ; M. COLA, Auteur ; S. PLATE, Auteur ; L. YANKOWITZ, Auteur ; J. PANDEY, Auteur ; Robert T. SCHULTZ, Auteur ; Julia PARISH-MORRIS, Auteur . - 14 p.
Langues : Anglais (eng)
in Molecular Autism > 10 (2019) . - 14 p.
Mots-clés : Autism spectrum disorder Mentalizing Narratives Natural language processing Sex differences Social cognition Storytelling Word choice Index. décimale : PER Périodiques Résumé : Background: Narrative abilities are linked to social impairment in autism spectrum disorder (ASD), such that reductions in words about cognitive processes (e.g., think, know) are thought to reflect underlying deficits in social cognition, including Theory of Mind. However, research suggests that typically developing (TD) boys and girls tell narratives in sex-specific ways, including differential reliance on cognitive process words. Given that most studies of narration in ASD have been conducted in predominantly male samples, it is possible that prior results showing reduced cognitive processing language in ASD may not generalize to autistic girls. To answer this question, we measured the relative frequency of two kinds of words in stories told by autistic girls and boys: nouns (words that indicate object-oriented storytelling) and cognitive process words (words like think and know that indicate mentalizing or attention to other peoples' internal states). Methods: One hundred two verbally fluent school-aged children [girls with ASD (N = 21) and TD (N = 19), and boys with ASD (N = 41) and TD (N = 21)] were matched on age, IQ, and maternal education. Children told a story from a sequence of pictures, and word frequencies (nouns, cognitive process words) were compared. Results: Autistic children of both sexes consistently produced a greater number of nouns than TD controls, indicating object-focused storytelling. There were no sex differences in cognitive process word use in the TD group, but autistic girls produced significantly more cognitive process words than autistic boys, despite comparable autism symptom severity. Thus, autistic girls showed a unique narrative profile that overlapped with autistic boys and typical girls/boys. Noun use correlated significantly with parent reports of social symptom severity in all groups, but cognitive process word use correlated with social ability in boys only. Conclusion: This study extends prior research on autistic children's storytelling by measuring sex differences in the narratives of a relatively large, well-matched sample of children with and without ASD. Importantly, prior research showing that autistic children use fewer cognitive process words is true for boys only, while object-focused language is a sex-neutral linguistic marker of ASD. These findings suggest that sex-sensitive screening and diagnostic methods-preferably using objective metrics like natural language processing-may be helpful for identifying autistic girls, and could guide the development of future personalized treatment strategies. En ligne : http://dx.doi.org/10.1186/s13229-019-0268-2 Permalink : https://www.cra-rhone-alpes.org/cid/opac_css/index.php?lvl=notice_display&id=398