Natural language processing in at-risk mental states: enhancing the assessment of thought disorders and psychotic traits with semantic dynamics and graph theory.

Autor:	Argolo F; Laboratório de Neurociencias (LIM 27), Instituto de Psiquiatria, Hospital das Clínicas HCFMUSP, Faculdade de Medicina, Universidade de São Paulo, São Paulo, SP, Brazil., Ramos WHP; Departamento de Estatística, Universidade Federal do Paraná, Curitiba, PR, Brazil., Mota NB; Instituto de Psiquiatria (IPUB), Departamento de Psiquiatria e Medicina Legal, Universidade Federal do Rio de Janeiro (UFRJ), Rio de Janeiro, RJ, Brazil. Research department at Motrix Lab - Motrix, Rio de Janeiro, RJ, Brazil., Gondim JM; Instituto de Computação, Universidade Federal da Bahia, Salvador, BA, Brazil., Lopes-Rocha AC; Laboratório de Neurociencias (LIM 27), Instituto de Psiquiatria, Hospital das Clínicas HCFMUSP, Faculdade de Medicina, Universidade de São Paulo, São Paulo, SP, Brazil., Andrade JC; Laboratório de Neurociencias (LIM 27), Instituto de Psiquiatria, Hospital das Clínicas HCFMUSP, Faculdade de Medicina, Universidade de São Paulo, São Paulo, SP, Brazil., van de Bilt MT; Laboratório de Neurociencias (LIM 27), Instituto de Psiquiatria, Hospital das Clínicas HCFMUSP, Faculdade de Medicina, Universidade de São Paulo, São Paulo, SP, Brazil. Instituto Nacional de Biomarcadores em Neuropsiquiatria (INBION), Conselho Nacional de Desenvolvimento Científico e Tecnológico, Brazil., de Jesus LP; Laboratório de Neurociencias (LIM 27), Instituto de Psiquiatria, Hospital das Clínicas HCFMUSP, Faculdade de Medicina, Universidade de São Paulo, São Paulo, SP, Brazil., Jafet A; Laboratório de Neurociencias (LIM 27), Instituto de Psiquiatria, Hospital das Clínicas HCFMUSP, Faculdade de Medicina, Universidade de São Paulo, São Paulo, SP, Brazil., Cecchi G; IBM T. J. Watson Research Center, Yorktown Heights, NY, USA., Gattaz WF; Laboratório de Neurociencias (LIM 27), Instituto de Psiquiatria, Hospital das Clínicas HCFMUSP, Faculdade de Medicina, Universidade de São Paulo, São Paulo, SP, Brazil. Instituto Nacional de Biomarcadores em Neuropsiquiatria (INBION), Conselho Nacional de Desenvolvimento Científico e Tecnológico, Brazil., Corcoran CM; Icahn School of Medicine at Mount Sinai New York, NY, USA. James J. Peters VA Medical Center Bronx, NY, USA., Ara A; Departamento de Estatística, Universidade Federal do Paraná, Curitiba, PR, Brazil., Loch AA; Laboratório de Neurociencias (LIM 27), Instituto de Psiquiatria, Hospital das Clínicas HCFMUSP, Faculdade de Medicina, Universidade de São Paulo, São Paulo, SP, Brazil. Instituto Nacional de Biomarcadores em Neuropsiquiatria (INBION), Conselho Nacional de Desenvolvimento Científico e Tecnológico, Brazil.
Jazyk:	angličtina
Zdroj:	Revista brasileira de psiquiatria (Sao Paulo, Brazil : 1999) [Braz J Psychiatry] 2024 Jul 29. Date of Electronic Publication: 2024 Jul 29.
DOI:	10.47626/1516-4446-2023-3419
Abstrakt:	Objective: Verbal communication has key information for mental health evaluation. Researchers have linked psychopathology phenomena to some of their counterparts in natural-language-processing (NLP). We study the characterization of subtle impairments presented in early stages of psychosis, developing new analysis techniques and a comprehensive map associating NLP features with the full range of clinical presentation. Methods: We used NLP to assess elicited and free-speech of 60 individuals in at-risk-mental-states (ARMS) and 73 controls, screened from 4,500 quota-sampled Portuguese speaking citizens in Sao Paulo, Brazil. Psychotic symptoms were independently assessed with Structured-Interview-for-Psychosis-Risk-Syndromes (SIPS). Speech features (e.g.sentiments, semantic coherence), including novel ones, were correlated with psychotic traits (Spearman's-ρ) and ARMS status (general linear models and machine-learning ensembles). Results: NLP features were informative inputs for classification, which presented 86% balanced accuracy. The NLP features brought forth (e.g. Semantic laminarity as 'perseveration', Semantic recurrence time as 'circumstantiality', average centrality in word repetition graphs) carried most information and also presented direct correlations with psychotic symptoms. Out of the standard measures, grammatical tagging (e.g. use of adjectives) was the most relevant. Conclusion: Subtle speech impairments can be grasped by sensitive methods and used for ARMS screening. We sketch a blueprint for speech-based evaluation, pairing features to standard thought disorder psychometric items. Competing Interests: FA has provided consulting services and developed technology for private companies. NBM works at Mobile Brain, an Education and Health Tech startup, and has been a consultant to Boehringer Ingelheim. JMG works at mobile Brain and has provided consulting for developing machine learning models for private companies. The other authors report no conflicts of interest.
Databáze:	MEDLINE
Externí odkaz:	Zobrazit plný text záznamu