Collection

Language Resources for Clinical Linguistics

In the last decades, a rich and growing body of research has explored the linguistic correlates of medical conditions. While great advances have been made in the profiling of “pathological” speech, several theoretical and methodological issues are still unsolved. What do corpus-based studies of speech and language disorders bring to the table in terms of description and theory, with respect to standardized testing? Can “good practices” be established for the elicitation, collection, transcription, and annotation of clinical linguistic datasets? Since patients’ recordings, transcripts, and written productions are “special categories of personal data” subjected to stringent data-protection safeguards, is data reusability possible?

As a matter of fact, clinical linguistics experiences a chronic lack of reliable data and a recurrent scattering of efforts in resource collection. Furthermore, the approval by ethics committees usually limits the shareability of clinically-valid corpora. Thus, research groups can only develop small linguistic resources, usable for specific purposes explicitly authorized by the research ethics boards.

In this framework, this special issue focuses on the methodologies for collecting linguistic resources – corpora, lexica, database, ontologies – and developing (computational) tools for the annotation of atypical language due to neurodevelopmental or acquired disabilities (e.g., Developmental Language Disorder, Autism, Aphasia, Mild Cognitive Impairment and Dementia, Parkinson's Disease & Parkinsonism, Motor Neuron Disease).

Topics of Interest

The editors invite original, unpublished contributions on topics including but not limited to:

Corpus-based research in speech-language pathology: acquisition, creation, annotation.

Identification and use of language resources for clinical applications (e.g., screening, diagnosis, monitoring, and phenotyping of language-related disorders).

Design of annotation tools for the analysis of ‘atypical’ verbal productions.

Legal and ethical aspects of language technology: good practices for the collection, treatment, storage, sharing, and dissemination of linguistic data from healthcare settings.

Editors

  • Gloria Gagliardi

    Gloria Gagliardi is currently senior Assistant Professor at the University of Bologna, Department of Classical Philology and Italian Studies. Graduated in “Linguistica Italiana e Civiltà Letterarie” (Modern Philology) with honours in 2010, she obtained her PhD degree in Linguistics in 2014 defending a thesis titled: “Validazione dell’ontologia dell’azione IMAGACT per lo studio e la diagnosi del Mild Cognitive Impairment” (“Validation of IMAGACT ontology of action for the study and diagnosis of Mild Cognitive Impairment”). She has concentrated her professional experience on the study of the Italian Language, specifically in its spoken fo

  • Marta Maffia

    Marta Maffia currently works at the Department of Literary, Linguistics and Comparative Studies at the University of Naples “L’Orientale” in Italy.

Articles

Articles will be displayed here once they are published.