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

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.

Stimulation of microvesicle secretion in Trichomonas vaginalis.

Trichomonas vaginalis is an extracellular flagellate protozoan and the etiological agent of human trichomoniasis, a sexually transmitted infection (STI) with a high incidence. Several reports have shown that this protozoan releases microvesicles into the culture medium, which show high potential in modulating cell-to-cell communication and the host response to infections. However, the biogenesis of these vesicles has not been analyzed in detail. In the present study, high-resolution ion scanning microscopy (SEM) and transmission electron microscopy (TEM) were used to analyze the surface of control cells and cells incubated in the presence of Ca2+ alone or with A 23187 calcium ionophore. Two different strains of T. vaginalis were analyzed. Most control cells displayed relatively smooth surfaces, whereas cells incubated with Ca2+ had many surface projections of variable shape and size (from 40 nm to around 1 µm). Quantitative analyses were performed directly in the scanning electron microscope and showed a significant increase in the number of cells with surface projections after incubation in the presence of calcium. TEM showed that treated cells presented several cytoplasmic multivesicular structures, suggesting membrane fusion and exosomes in the extracellular medium. The amount and size of the released vesicles were quantitatively analyzed using light scattering and TEM on negatively stained samples. The observations show that incubation of both parasite strains in the presence of Ca2+ significantly increased the release of microvesicles into the extracellular medium in a time-dependent process. Sequential incubation in the presence of Ca2+ and the calcium ionophore A23187 increases the presence of vesicles on the parasite surface only at a short incubation time (5 min). Transmission electron microscopy showed that at least part of the vesicles are originated from cytoplasmic multivesicular structures. This information contributes to a better understanding of the biogenesis of extracellular vesicles secreted by T. vaginalis.

Cannabis use influences disorganized symptoms severity but not transition in a cohort of non-help-seeking individuals at-risk for psychosis from São Paulo, Brazil.

BACKGROUND: Cannabis use is associated with an increased risk of developing a psychotic disorder. However, in individuals with at-risk mental states for psychosis (ARMS) this association is not clear, as well as the impact of cannabis use on symptom severity. The objective of this study was to evaluate the association of cannabis use patterns and ARMS risk status, transition to psychotic and psychiatric disorders, and psychopathology. METHOD: A sample of 109 ARMS and 197 control individuals was drawn from the general population. Lifetime, maximum and current amount of cannabis use were assessed with the South Westminster modified questionnaire. Participants were followed-up for a mean of 2.5 years and reassessed for transition to any psychiatric disorder. RESULTS: There were no differences between ARMS and controls regarding lifetime use, current amount of use, or maximum amount of cannabis use. There were also no differences between those who transitioned to a psychiatric disorder and those who did not regarding cannabis use variables. In ARMS individuals, cannabis use was significantly related to disorganization symptoms. CONCLUSION: The results of this study suggest that cannabis plays a role in the psychopathology of ARMS individuals, leading to more severe symptomatology.

Plasma levels of neurotrophin 4/5, NGF and pro-BDNF influence transition to mental disorders in a sample of individuals at ultra-high risk for psychosis.

BACKGROUND: Neurotrophins (NTs) and their precursors (pro-NTs) are polypeptides with important roles in neuronal development, differentiation, growth, survival and plasticity, as well as apoptosis and neuronal death. Imbalance in NT levels were observed in schizophrenia spectrum disorders, but evidence in ultra-high risk for psychosis (UHR) samples is scarce. METHODS: A naturalistic sample of 87 non-help-seeking UHR subjects and 55 healthy controls was drawn from the general population. Blood samples were collected and NT-3, NT-4/5, BDNF, pro-BDNF, NGF, pro-NGF were analyzed through enzyme linked immunosorbent assay (ELISA). Information on cannabis and tobacco use was also collected. Logistic regression models and path analysis were used to control for confounders (tobacco, age, cannabis use). RESULTS: NT-4/5 was significantly decreased, and pro-BDNF was significantly increased in UHR individuals compared to controls. Cannabis use and higher NGF levels were significantly related to transition to psychiatric disorders among UHR subjects. Increased pro-BDNF and decreased NT-4/5 influenced transition by the mediation of perceptual abnormalities. CONCLUSIONS: Our study shows for the first time that NTs are altered in UHR compared to healthy control individuals, and that they can be a predictor of transition to psychiatric illnesses in this population. Future studies should employ larger naturalistic samples to confirm the findings.

Detecting at-risk mental states for psychosis (ARMS) using machine learning ensembles and facial features.

AIMS: Our study aimed to develop a machine learning ensemble to distinguish "at-risk mental states for psychosis" (ARMS) subjects from control individuals from the general population based on facial data extracted from video-recordings. METHODS: 58 non-help-seeking medication-naïve ARMS and 70 healthy subjects were screened from a general population sample. At-risk status was assessed with the Structured Interview for Prodromal Syndromes (SIPS), and "Subject's Overview" section was filmed (5-10 min). Several features were extracted, e.g., eye and mouth aspect ratio, Euler angles, coordinates from 51 facial landmarks. This elicited 649 facial features, which were further selected using Gradient Boosting Machines (AdaBoost combined with Random Forests). Data was split in 70/30 for training, and Monte Carlo cross validation was used. RESULTS: Final model reached 83 % of mean F1-score, and balanced accuracy of 85 %. Mean area under the curve for the receiver operator curve classifier was 93 %. Convergent validity testing showed that two features included in the model were significantly correlated with Avolition (SIPS N2 item) and expression of emotion (SIPS N3 item). CONCLUSION: Our model capitalized on short video-recordings from individuals recruited from the general population, effectively distinguishing between ARMS and controls. Results are encouraging for large-screening purposes in low-resource settings.

Gesticulation in individuals with at risk mental states for psychosis.

Nonverbal communication (NVC) is a complex behavior that involves different modalities that are impaired in the schizophrenia spectrum, including gesticulation. However, there are few studies that evaluate it in individuals with at-risk mental states (ARMS) for psychosis, mostly in developed countries. Given our prior findings of reduced movement during speech seen in Brazilian individuals with ARMS, we now aim to determine if this can be accounted for by reduced gesticulation behavior. Fifty-six medication-naïve ARMS and 64 healthy controls were filmed during speech tasks. The frequency of specifically coded gestures across four categories (and self-stimulatory behaviors) were compared between groups and tested for correlations with prodromal symptoms of the Structured Interview for Prodromal Syndromes (SIPS) and with the variables previously published. ARMS individuals showed a reduction in one gesture category, but it did not survive Bonferroni's correction. Gesture frequency was negatively correlated with prodromal symptoms and positively correlated with the variables of the amount of movement previously analyzed. The lack of significant differences between ARMS and control contradicts literature findings in other cultural context, in which a reduction is usually seen in at-risk individuals. However, gesture frequency might be a visual proxy of prodromal symptoms, and of other movement abnormalities. Results show the importance of analyzing NVC in ARMS and of considering different cultural and sociodemographic contexts in the search for markers of these states.

A dopamine receptor D2 genetic polymorphism associated with transition to mental disorders in a cohort of individuals with at-risk mental state for psychosis.

OBJECTIVES: To test the association of 45 single nucleotide polymorphisms (SNPs) with transition to psychiatric disorders in a cohort of individuals at ultrahigh risk (UHR) mental state for psychosis. METHODS: Through general population screening, 88 non-help-seeking UHR subjects and 130 healthy control individuals were genotyped for 45 SNPs related to psychosis. They were followed for a mean of 2.5 years, and conversion to psychotic and to general psychiatric disorders was assessed. Genotype frequencies between controls, converters, and non-converters were analyzed. RESULTS: There were no differences in sociodemographics between controls and UHR. Also, UHR converters and non-converters had no differences in their baseline symptoms scores. The dopamine receptor D2 gene (DRD2) SNP rs6277 was significantly more common among UHR who transitioned to psychosis (p < 0.001) and to UHR who transitioned to any psychiatric disorders (p = 0.001) when compared to UHR who did not transition. The rs6277 T allele was related to psychiatric morbidity in a dose-response fashion, being significantly more frequent in UHR converters than UHR non-converters and control subjects (p = 0.003). CONCLUSION: Our findings suggest that rs6277 could potentially constitute a genetic marker of transition to psychiatric disorders in subjects with at-risk mental states, warranting further investigation in larger samples.

A dopamine receptor D2 genetic polymorphism associated with transition to mental disorders in a cohort of individuals with at-risk mental state for psychosis

Objectives: To test the association of 45 single nucleotide polymorphisms (SNPs) with transition to psychiatric disorders in a cohort of individuals at ultrahigh risk (UHR) mental state for psychosis. Methods: Through general population screening, 88 non-help-seeking UHR subjects and 130 healthy control individuals were genotyped for 45 SNPs related to psychosis. They were followed for a mean of 2.5 years, and conversion to psychotic and to general psychiatric disorders was assessed. Genotype frequencies between controls, converters, and non-converters were analyzed. Results: There were no differences in sociodemographics between controls and UHR. Also, UHR converters and non-converters had no differences in their baseline symptoms scores. The dopamine receptor D2 gene (DRD2) SNP rs6277 was significantly more common among UHR who transitioned to psychosis (p < 0.001) and to UHR who transitioned to any psychiatric disorders (p = 0.001) when compared to UHR who did not transition. The rs6277 T allele was related to psychiatric morbidity in a dose-response fashion, being significantly more frequent in UHR converters than UHR non-converters and control subjects (p = 0.003). Conclusion: Our findings suggest that rs6277 could potentially constitute a genetic marker of transition to psychiatric disorders in subjects with at-risk mental states, warranting further investigation in larger samples.

Motion energy analysis during speech tasks in medication-naïve individuals with at-risk mental states for psychosis.

Movement abnormalities are commonly observed in schizophrenia and at-risk mental states (ARMS) for psychosis. They are usually detected with clinical interviews, such that automated analysis would enhance assessment. Our aim was to use motion energy analysis (MEA) to assess movement during free-speech videos in ARMS and control individuals, and to investigate associations between movement metrics and negative and positive symptoms. Thirty-two medication-naïve ARMS and forty-six healthy control individuals were filmed during speech tasks. Footages were analyzed using MEA software, which assesses movement by differences in pixels frame-by-frame. Two regions of interest were defined-head and torso-and mean amplitude, frequency, and coefficient of variability of movements for them were obtained. These metrics were correlated with the Structured Interview for Prodromal Syndromes (SIPS) symptoms, and with the risk of conversion to psychosis-inferred with the SIPS risk calculator. ARMS individuals had significantly lower mean amplitude of head movement and higher coefficients of movement variability for both head and torso, compared to controls. Higher coefficient of variability was related to higher risk of conversion. Negative correlations were seen between frequency of movement and most SIPS negative symptoms. All positive symptoms were correlated with at least one movement variable. Movement abnormalities could be automatically detected in medication-naïve ARMS subjects by means of a motion energy analysis software. Significant associations of movement metrics with symptoms were found, supporting the importance of movement analysis in ARMS. This could be a potentially important tool for early diagnosis, intervention, and outcome prediction.

Use of a Bayesian Network Model to predict psychiatric illness in individuals with 'at risk mental states' from a general population cohort.

The 'at risk mental state' (ARMS) paradigm has been introduced in psychiatry to study prodromal phases of schizophrenia. With time it was seen that the ARMS state can also precede mental disorders other than schizophrenia, such as depression and anxiety. However, several problems hamper the paradigm's use in preventative medicine, such as varying transition rates across studies, the use of non-naturalistic samples, and the multifactorial nature of psychiatric disorders. To strengthen ARMS predictive power, there is a need for a holistic model incorporating-in an unbiased fashion-the small-effect factors that cause mental disorders. Bayesian networks, a probabilistic graphical model, was used in a populational cohort of 83 ARMS individuals to predict conversion to psychiatric illness. Nine predictors-including state, trait, biological and environmental factors-were inputted. Dopamine receptor 2 polymorphism, high private religiosity, and childhood trauma remained in the final model, which reached an 85.51% (SD = 0.1190) accuracy level in predicting conversion. This is the first time a robust model was produced with Bayesian networks to predict psychiatric illness among at risk individuals from the general population. This could be an important tool to strengthen predictive measures in psychiatry which should be replicated in larger samples to provide the model further learning.