Language as a biomarker for psychosis: A natural language processing approach

SCHIZOPHRENIA RESEARCH

卷:226

Article

Corcoran, Cheryl M.¹  Mittal, Vijay A.²  Bearden, Carrie E.^3,9,10  Gur, Raquel E.⁴  Hitczenko, Kasia⁸  Bilgrami, Zarina¹  Savic, Aleksandar⁵  Cecchi, Guillermo A.⁶  Wolff, Phillip⁷ 

[1] Icahn Sch Med Mt Sinai, Dept Psychiat, New York, NY 10029 USA

[2] Northwestern Univ, Dept Psychol, Evanston, IL USA

[3] Univ Calif Los Angeles, Dept Psychiat & Biobehav Sci, Los Angeles, CA 90024 USA

[4] Dept Psychiat, Neuropsychiat Div, Brain Behav Lab, Philadelphia, PA 19104 USA

[5] Univ Psychiat Hosp Vrapce, Dept Diagnost & Intens Care, Zagreb, Croatia

[6] IBM Corp, TJ Watson Res Ctr, Computat Biol Ctr Neurosci, Yorktown Hts, NY USA

[7] Emory Univ, Dept Psychol, Atlanta, GA 30322 USA

[8] Northwestern Univ, Dept Linguist, Evanston, IL USA

[9] Univ Calif Los Angeles, Semel Inst Neurosci & Human Behav, Brain Res Inst, Dept Psychol, Los Angeles, CA 90024 USA

[10] Univ Calif Los Angeles, Dept Psychol, Los Angeles, CA 90024 USA

关键词: Psychosis;    Automated language analysis;    Natural language processing;    Machine learning;    Semantic coherence;    Discourse coherence;    Referential coherence;    Semantic density;    Latent semantic analysis;    Digital phenotyping;    Psychosis risk;    Clinical high risk;    Ultra high risk;    Schizophrenia;   

DOI  :  10.1016/j.schres.2020.04.032

来源: Elsevier

【 摘 要 】

Human ratings of conceptual disorganization, poverty of content, referential cohesion and illogical thinking have been shown to predict psychosis onset in prospective clinical high risk (CHR) cohort studies. The potential value of linguistic biomarkers has been significantly magnified, however, by recent advances in natural language processing (NLP) and machine learning (ML). Such methodologies allow for the rapid and objective measurement of language features, many of which are not easily recognized by human raters. Here we review the key findings on language production disturbance in psychosis. We also describe recent advances in the computational methods used to analyze language data, including methods for the automatic measurement of discourse coherence, syntactic complexity, poverty of content, referential coherence, and metaphorical language. Linguistic biomarkers of psychosis risk are now undergoing cross-validation, with attention to harmonization of methods. Future directions in extended CHR networks include studies of sources of variance, and combination with other promising biomarkers of psychosis risk, such as cognitive and sensory processing impairments likely to be related to language. Implications for the broader study of social communication, including reciprocal prosody, face expression and gesture, are discussed. (C) 2020 Elsevier B.V. All rights reserved.

【 授权许可】

Free