Assessing dimensions of thought disorder with large language models: The tradeoff of accuracy and consistency.

Natural Language Processing (NLP) methods have shown promise for the assessment of formal thought disorder, a hallmark feature of schizophrenia in which disturbances to the structure, organization, or coherence of thought can manifest as disordered or incoherent speech. We investigated the suitability of modern Large Language Models (LLMs - e.g., GPT-3.5, GPT-4, and Llama 3) to predict expert-generated ratings for three dimensions of thought disorder (coherence, content, and tangentiality) assigned to speech samples collected from both patients with a diagnosis of schizophrenia (n = 26) and healthy control participants (n = 25). In addition to (1) evaluating the accuracy of LLM-generated ratings relative to human experts, we also (2) investigated the degree to which the LLMs produced consistent ratings across multiple trials, and we (3) sought to understand the factors that impacted the consistency of LLM-generated output. We found that machine-generated ratings of the level of thought disorder in speech matched favorably those of expert humans, and we identified a tradeoff between accuracy and consistency in LLM ratings. Unlike traditional NLP methods, LLMs were not always consistent in their predictions, but these inconsistencies could be mitigated with careful parameter selection and ensemble methods. We discuss implications for NLP-based assessment of thought disorder and provide recommendations of best practices for integrating these methods in the field of psychiatry.

Early Intervention in the Treatment of Psychosis.

This article aims to: (1) describe the evolution of first episode of psychosis (FEP) approaches; (2) define a model of multidisciplinary care; (3) identify challenges and limitations; (4) discuss the unique challenges for those first experiencing psychosis; (5) identify strategies to expand early psychosis interventions. The authors take the medical standpoint and use the differential diagnosis and initial medical work-up as a context for assessment. The remainder of the article will be focused on treatment of FEP in those with schizophrenia-spectrum disorders.

Schizophrenia diagnosis using the GRU-layer's alpha-EEG rhythm's dependability.

Verifying schizophrenia (SZ) can be assisted by deep learning techniques and patterns in brain activity observed in alpha-EEG recordings. The suggested research provides evidence of the reliability of alpha-EEG rhythm in a Gated-Recurrent-Unit-based deep-learning model for investigating SZ. This study suggests Rudiment Densely-Coupled Convolutional Gated Recurrent Unit (RDCGRU) for the various EEG-rhythm-based (gamma, beta, alpha, theta, and delta) diagnoses of SZ. The model includes multiple 1-D-Convolution (Con-1-D) folds with steps greater than 1, which enables the model to programmatically and effectively learn how to reduce the incoming signal. The Con-1-D layers and numerous Gated Recurrent Unit (GRU) layers comprise the Exponential-Linear-Unit activation function. This powerful activation function facilitates in-deep-network training and improves classification performance. The Densely-Coupled Convolutional Gated Recurrent Unit (DCGRU) layers enable RDCGRU to address the training accuracy loss brought on by vanishing or exploding gradients, and this might make it possible to develop intense, deep versions of RDCGRU for more complex problems. The sigmoid activation function is implemented in the digital (binary) classifier's output nodes. The RDCGRU deep learning model attained the most excellent accuracy, 88.88 %, with alpha-EEG rhythm. The research achievements: The RDCGRU deep learning model's GRU cells responded superiorly to the alpha-EEG rhythm in EEG-based verification of SZ.

Impact of early risk factors on schizophrenia risk and age of diagnosis: A Danish population-based register study.

BACKGROUND: While several risk factors for schizophrenia have been identified, their individual impacts are rather small. The relative independent and cumulative impacts of multiple risk factors on disease risk and age of onset warrant further investigation. STUDY DESIGN: We conducted a register-based case-control study including all individuals receiving a schizophrenia spectrum disorder in Denmark from 1973 to 2018 (N = 29,142), and a healthy control sample matched 5:1 on age, sex, and parental socioeconomic status (N = 136,387). Register data included parental history of psychiatric illness, birth weight, gestational age, season of birth, population density of birthplace, immigration, paternal age, and Apgar scores. Data were analysed using logistic regression and machine learning. RESULTS: Parental history of psychiatric illness (OR = 2.32 [95%CI 2.21-2.43]), high paternal age (OR = 1.30 [1.16-1.45]), and low birth weight (OR = 1.28 [1.16-1.41]) increased the odds of belonging to the patient group. In contrast, being a second-generation immigrant (OR = 0.65 [0.61-0.69]) and high population density of the birthplace (OR = 0.92 [0.89-0.96]) decreased the odds. The findings were supported by a decision tree analysis where parental history, paternal age, and birth weight contributed most to diagnostic classification (ACCtest = 0.69, AUCtest = 0.59, p < 0.001). Twenty percent of patients were child-onset cases. Here, female sex (OR = 1.82 [1.69-1.97]) and parental psychiatric illness (OR = 1.62 [1.49-1.77]) increased the odds of receiving the diagnosis <18 years. CONCLUSION: Multiple early factors contribute independently to a higher psychosis risk, suggesting cumulative effects leading to symptom onset. Routine assessments of the most influential risk factors could be incorporated into clinical practise. Being female increased the risk of diagnosis during childhood, suggesting sex differences in the developmental trajectories of the disorder.

The prognosis of schizophrenia.

Psychotic and negative symptoms are core features of schizophrenia but knowledge on the long-term course of these symptoms is lacking. While antipsychotics improve psychotic symptoms, they have limited effect on negative symptoms. Specialized early interventions like OPUS show great effects while ongoing but long-term follow-up indicates no lasting disease-modifying effects. 18% of patients achieved clinical recovery, and 40% obtained symptom remission after 20 years, but high mortality rates and continuous negative symptoms constitute an unmet treatment need, as argued in this review.

Validity of handgrip strength for assessing cognition and psychotic symptoms in hospitalized patients with stable schizophrenia.

BACKGROUND AND OBJECTIVES: A correlation between low handgrip strength (HGS), HGS asymmetry, and low cognitive performance has been demonstrated. However, it remains unclear whether low HGS is associated with psychotic symptoms and whether HGS asymmetry is associated with cognitive and psychotic symptoms in hospitalized patients with schizophrenia. This study aimed to investigate the validity of HGS as a measure for assessing cognition and psychotic symptoms in hospitalized patients with stable schizophrenia. METHODS: A total of 235 inpatients with stable schizophrenia were recruited between August 1, 2023, and August 31, 2023. The highest HGS values from three tests on the dominant hand were used to determine low HGS (male < 28 kg, female < 18 kg), and HGS asymmetry was identified when the non-dominant HGS/dominant HGS ratio was outside 0.9-1.1. Cognition and psychotic symptoms were assessed using the Chinese Montreal Cognitive Assessment (MoCA-C) and Positive and Negative Syndrome Scale (PANSS). Generalized linear model analyses examined the relationship between HGS and scale scores. RESULTS: Covariate-adjusted generalized linear models confirmed a strong association between low HGS alone and the MoCA-C score (OR = 0.819, 95% CI = 0.710‒0.945, p = 0.006) and PANSS score (OR = 1.113, 95% CI = 1.036‒1.239, p = 0.006). Similarly, the combination of low and asymmetric HGS was strongly associated with both MoCA-C (OR = 0.748, 95% CI = 0.653‒0.857, p<0.001) and PANSS scores (OR = 1.118, 95% CI = 1.032‒1.211, p = 0.006). CONCLUSIONS: The results suggest that hospitalized patients with schizophrenia and low HGS, with or without asymmetry, are likely to have lower MoCA-C scores and higher PANSS scores. Screening stable schizophrenia patients with low HGS, with or without asymmetry, could be a valuable and straightforward approach to identifying those with lower cognition and severe psychotic symptoms.

Joint multi-site domain adaptation and multi-modality feature selection for the diagnosis of psychiatric disorders.

Identifying biomarkers for computer-aided diagnosis (CAD) is crucial for early intervention of psychiatric disorders. Multi-site data have been utilized to increase the sample size and improve statistical power, while multi-modality classification offers significant advantages over traditional single-modality based approaches for diagnosing psychiatric disorders. However, inter-site heterogeneity and intra-modality heterogeneity present challenges to multi-site and multi-modality based classification. In this paper, brain functional and structural networks (BFNs/BSNs) from multiple sites were constructed to establish a joint multi-site multi-modality framework for psychiatric diagnosis. To do this we developed a hypergraph based multi-source domain adaptation (HMSDA) which allowed us to transform source domain subjects into a target domain. A local ordinal structure based multi-task feature selection (LOSMFS) approach was developed by integrating the transformed functional and structural connections (FCs/SCs). The effectiveness of our method was validated by evaluating diagnosis of both schizophrenia (SZ) and autism spectrum disorder (ASD). The proposed method obtained accuracies of 92.2 %±2.22 % and 84.8 %±2.68 % for the diagnosis of SZ and ASD, respectively. We also compared with 6 DA, 10 multi-modality feature selection, and 8 multi-site and multi-modality methods. Results showed the proposed HMSDA+LOSMFS effectively integrated multi-site and multi-modality data to enhance psychiatric diagnosis and identify disorder-specific diagnostic brain connections.

Borderlines of psychosis - nosological propositions of Polish psychiatrists of the interwar period. / Pogranicza psychozy ­ nozologiczne propozycje polskich psychiatrów dwudziestolecia miedzywojennego.

This paper examines nosological categories relating to borderlines between psychosis and other clinical categories, introduced by Polish psychiatrists in the interwar period. In the United States, the discussion about the borderline between neuroses and psychoses was urged by the 1938 article by psychoanalyst Adolph Stern. In Poland, nosological categories regarding the borderline between neuroses and psychoses were proposed by Adam Wizel, Maurycy Bornsztajn, Jan Nelken, and Wladyslaw Matecki. Wizel coined the term 'underdeveloped schizophrenia', Bornsztajn introduced 'schizothymia reactiva' and 'hypochondriac (somatopsychic) schizophrenia', Nelken described 'mild schizophrenia', first introduced by Moscow psychiatric school of Rosenstein, and Matecki presented the category of neurosis-like (pseudo-neurotic) schizophrenia. Additionally, Julian Dretler, after studying the borderline between schizophrenia and manic-depressive psychosis, coined the term 'mixed psychosis' and expressed conviction that it is an independent nosological entity. Like in the United States, the majority of Polish pioneers of the nosological studies of borderline cases were influenced by psychoanalysis. As a consequence of World War II and the new regime, which forced dialectical materialism and Pavlovism as an official ideology of psychiatry and condemned psychoanalysis, the categories presented in the article became forgotten and have not impacted Polish psychiatric nosology.

Altered lipidomics biosignatures in schizophrenia: A systematic review.

Multiomics approaches have significantly aided the identification of molecular signatures in complex neuropsychiatric disorders. Lipidomics, one of the newest additions in the -omics family, sheds light on lipid profiles and is an emerging methodological tool to study schizophrenia pathobiology, as lipid dysregulation has been repeatedly observed in schizophrenia. In this review, we performed a detailed literature search for lipidomics studies in schizophrenia. Following elaborate inclusion/exclusion criteria, we focused on human studies in schizophrenia and schizophrenia-related diagnoses in brain and blood specimens, including serum plasma, platelets and red blood cells. Eighteen studies fulfilled our inclusion criteria, of which five were conducted in the brain, 12 in peripheral material and one in both. Here, we first provide background on lipidomics and the main lipid categories addressed, review in detail the included literature and look for common lipidomics patterns in brain and the periphery that emerge from these studies. Furthermore, we highlight current limitations in schizophrenia lipidomics research and underline the need for following up on lipidomics results with complementary molecular approaches.

Efficacy and tolerability of antipsychotic polypharmacy for schizophrenia spectrum disorders. A systematic review and meta-analysis of individual patient data.

BACKGROUND: Antipsychotic polypharmacy (APP) is frequently prescribed for schizophrenia-spectrum disorders. Despite the inconsistent findings on efficacy, APP may be beneficial for subgroups of psychotic patients. This meta-analysis of individual patient data investigated moderators of efficacy and tolerability of APP in adult patients with schizophrenia-spectrum disorders. DESIGN: We searched PubMed, EMBASE, and the Cochrane Central Register of Randomized Trials until September 1, 2022, for randomized controlled trials comparing APP with antipsychotic monotherapy. We estimated the effects with a one-stage approach for patient-level moderators and a two-stage approach for study-level moderators, using (generalized) linear mixed-effects models. Primary outcome was treatment response, defined as a reduction of 25 % or more in the Positive and Negative Syndrome Scale (PANSS) score. Secondary outcomes were study discontinuation, and changes from baseline on the PANSS total score, its positive and negative symptom subscale scores, the Clinical Global Impressions Scale (CGI), and adverse effects. RESULTS: We obtained individual patient data from 10 studies (602 patients; 31 % of all possible patients) and included 599 patients in our analysis. A higher baseline PANSS total score increased the chance of a response to APP (OR = 1.41, 95 % CI 1.02; 1.94, p = 0.037 per 10-point increase in baseline PANSS total), mainly driven by baseline positive symptoms. The same applied to changes on the PANSS positive symptom subscale and the CGI severity scale. Extrapyramidal side effects increased significantly where first and second-generation antipsychotics were co-prescribed. Study discontinuation was comparable between both treatment arms. CONCLUSIONS: APP was effective in severely psychotic patients with high baseline PANSS total scores and predominantly positive symptoms. This effect must be weighed against potential adverse effects.

Immune-based Machine learning Prediction of Diagnosis and Illness State in Schizophrenia and Bipolar Disorder.

BACKGROUND: Schizophrenia and bipolar disorder frequently face significant delay in diagnosis, leading to being missed or misdiagnosed in early stages. Both disorders have also been associated with trait and state immune abnormalities. Recent machine learning-based studies have shown encouraging results using diagnostic biomarkers in predictive models, but few have focused on immune-based markers. Our main objective was to develop supervised machine learning models to predict diagnosis and illness state in schizophrenia and bipolar disorder using only a panel of peripheral kynurenine metabolites and cytokines. METHODS: The cross-sectional I-GIVE cohort included hospitalized acute bipolar patients (n = 205), stable bipolar outpatients (n = 116), hospitalized acute schizophrenia patients (n = 111), stable schizophrenia outpatients (n = 75) and healthy controls (n = 185). Serum kynurenine metabolites, namely tryptophan (TRP), kynurenine (KYN), kynurenic acid (KA), quinaldic acid (QUINA), xanthurenic acid (XA), quinolinic acid (QUINO) and picolinic acid (PICO) were quantified using liquid chromatography-tandem mass spectrometry (LC-MS/MS), while V-plex Human Cytokine Assays were used to measure cytokines (interleukin-6 (IL-6), IL-8, IL-17, IL-12/IL23-P40, tumor necrosis factor-alpha (TNF-ɑ), interferon-gamma (IFN-γ)). Supervised machine learning models were performed using JMP Pro 17.0.0. We compared a primary analysis using nested cross-validation to a split set as sensitivity analysis. Post-hoc, we re-ran the models using only the significant features to obtain the key markers. RESULTS: The models yielded a good Area Under the Curve (AUC) (0.804, Positive Prediction Value (PPV) = 86.95; Negative Prediction Value (NPV) = 54.61) for distinguishing all patients from controls. This implies that a positive test is highly accurate in identifying the patients, but a negative test is inconclusive. Both schizophrenia patients and bipolar patients could each be separated from controls with a good accuracy (SCZ AUC 0.824; BD AUC 0.802). Overall, increased levels of IL-6, TNF-ɑ and PICO and decreased levels of IFN-γ and QUINO were predictive for an individual being classified as a patient. Classification of acute versus stable patients reached a fair AUC of 0.713. The differentiation between schizophrenia and bipolar disorder yielded a poor AUC of 0.627. CONCLUSIONS: This study highlights the potential of using immune-based measures to build predictive classification models in schizophrenia and bipolar disorder, with IL-6, TNF-ɑ, IFN-γ, QUINO and PICO as key candidates. While machine learning models successfully distinguished schizophrenia and bipolar disorder from controls, the challenges in differentiating schizophrenic from bipolar patients likely reflect shared immunological pathways by the both disorders and confounding by a larger state-specific effect. Larger multi-centric studies and multi-domain models are needed to enhance reliability and translation into clinic.

Pre-stimulus EEG phase coherence predicts visual target detection failures in schizophrenia: A pilot study.

Impaired visual target detection is a common finding in schizophrenia that is linked to poor functional outcomes. However, the neural mechanisms that contribute to this deficit remain unclear. Recent research in healthy samples has identified relationships between the phase of pre-stimulus electroencephalographic (EEG) activity in the alpha band (8-12 Hz) or theta band (4-7 Hz) and the likelihood of visual target detection with and without attentional cueing, but these effects have not yet been explored in schizophrenia. We performed a study to investigate such effects in schizophrenia (n = 19) and healthy participants (n = 14), using a visual target detection task with attentional cues. We found significant relationships between pre-stimulus EEG phase properties and visual target detection in both groups, but also clear differences in the effects as a function of frequency, group, and attentional cueing. Alpha-band phase effects were relatively uniform across groups and conditions. By contrast, theta-band phase effects showed differences by group and attentional condition which could be consistent with attentional hyperfocusing in the schizophrenia group. Thus, our results elucidate a novel neural mechanism that may help to explain known impairments affecting both visual target detection and attention in schizophrenia.

Use of video-recordings of site-based interviews for quality assurance in a study of subjects with schizophrenia and persistent negative symptoms.

Site-independent ratings derived from audio-digital recordings of site-based interviews are often used for quality assurance monitoring to affirm ratings reliability in CNS clinical trials. The present study of subjects with schizophrenia and persistent negative symptoms used video instead of audio recordings of site-based interviews and thereby facilitated visual observation of the subject by the remote rater. "Paired" site-independent scores of the Positive and Negative Syndrome Scale (PANSS) and Brief Negative Symptom Scale (BNSS) were obtained from video-recordings of site-based interviews. The intraclass correlation between site-based and paired site-independent ratings was r = 0.839 for the total PANSS scores (n = 1006) and r = 0.871 for the total BNSS scores (n = 892); <5 % of paired scores deviated outside the acceptable confidence intervals. Ratings "outliers" were identified and remediated. We examined the pattern of paired scoring deviations for the BNSS, total PANSS, PANSS symptom subscales, and the Marder negative symptom factor. Each metric revealed a bidirectional pattern of scoring deviations such that mean site-based ratings were higher than site-independent ratings when symptom severity was high but lower than site-independent ratings when symptom severity was low. The pattern of bidirectional paired scoring deviations observed in this analysis has previously been noted in paired ratings analyses of subjects experiencing an acute exacerbation of psychosis in schizophrenia and major depressive disorder as well. The bidirectional pattern may reflect inherent differences between live ratings and remotely scored recorded ratings. This analysis affirms the utility of video-recordings of site-based ratings for surveillance in trials with subjects with schizophrenia and persistent negative symptoms.

[Controversial issues of diagnosis and theoretical concepts of schizophrenia in childhood]. / Diskussionnye voprosy diagnostiki i teoreticheskie modeli shizofrenii v detskom vozraste.

The article presents modern approaches to classification, presents debatable diagnostic issues, including the differences between domestic approaches to the diagnosis of schizophrenia in childhood from foreign taxonomies. The modern hypothesis of the etiological continuum of schizophrenic and autistic spectrum disorders is discussed, as well as clinical models of manifest stages of schizophrenia in childhood, with an emphasis on the influence of the age factor on the clinic, dynamics and prognosis of diseases.

[Schizophrenia spectrum disorders (aspects of psychopathology and clinical taxonomy)]. / Rasstroistva shizofrenicheskogo spektra (aspekty psikhopatologii i klinicheskaya sistematika).

OBJECTIVE: To create a new taxonomy of schizophrenia spectrum disorders (SSD) based on the comparability of the design of SSD and borderline states. MATERIAL AND METHODS: The total sample consists of 205 patients with an established diagnosis of SSD (F21; F25; F22 according to ICD-10) collected from studies of the department of borderline mental pathology and psychosomatic disorders of the Federal State Budgetary Institution Mental Health Research Center and the Department of Psychiatry and Psychosomatics of Moscow State Medical University in the period 2014 to 2024. Clinical, psychometric, statistical methods were used. RESULTS: A new two-level model of schizotypal personality disorder (STPD) has been developed: the first level is psychopathic-like disorders of the «Ferschroben¼ type; the second level are psychopathological disorders (positive, negative, etc.), appearing under their «mask¼, constituting a «tracing paper¼ of manifestations of schizophrenia «in miniature¼. The two-level psychopathological model of STPD is a complex clinical phenotype, including independent but overlapping phenotypic formations: psychopathic-like - the «Ferschroben¼ type; and basic - schizophreniform disorders. CONCLUSION: The clinical classification of schizophrenia spectrum disorders has been developed; pseudoneuroses and stress-induced disorders of the endogenous circle are considered in the aspect of the dynamics of STPD.

[Late-onset schizophrenia and schizophrenia-like psychosis with very late onset]. / Pozdnyaya shizofreniya i shizofrenopodobnyi psikhoz s ochen' pozdnim nachalom.

This review focuses on late-onset schizophrenia and schizophrenia-like psychosis with very late onset (VLOSLP) with focus on their psychopathologic, neuropsychologic, and neurobiologic aspects. A literature review on late-onset schizophrenia and VLOSLP was conducted based on publications from PubMed, Scopus, and Google Scholar databases up to December 2023. It may be noted that research into schizophrenia has largely focused on early-onset patients, and research into the mental health of older people has focused primarily on dementia and depression, with relatively little information on late-onset schizophrenia and VLOSLP. The nosology of late-onset functional psychoses is still poorly understood. There is currently no consensus on the diagnostic framework for psychosis labeled by the term VLOSLP. These deficiencies need to be addressed in order to understand the background of VLOSLP, the course and prognosis of the illness, and to develop successful management and treatment strategies for these patients, as older adults are more susceptible to the adverse effects of psychotropic medications. Therapy should be holistic, including not only medication but also psychotherapy, and the key role of caregivers of elderly schizophrenia patients should be taken into account. There should be judicious use of pharmacotherapy with an assessment of its risks and benefits.

The Interrelationships between Cytokines and Schizophrenia: A Systematic Review.

Schizophrenia (SCZ) imposes a significant burden on patients and their families because of its high prevalence rate and disabling nature. Given the lack of definitive conclusions regarding its pathogenesis, physicians heavily rely on patients' subjective symptom descriptions for diagnosis because reliable diagnostic biomarkers are currently unavailable. The role of the inflammatory response in the pathogenesis of SCZ has been supported by some studies. The findings of these studies showed abnormal changes in the levels of inflammatory factors, such as cytokines (CKs), in both peripheral blood and cerebrospinal fluid (CSF) among individuals affected by SCZ. The findings imply that inflammatory factors could potentially function as risk indicators for the onset of SCZ. Consequently, researchers have directed their attention towards investigating the potential utility of CKs as viable biomarkers for diagnosing SCZ. Extracellular vesicles (EVs) containing disease-specific components exhibit remarkable stability and abundance, making them promising candidates for biomarker discovery across various diseases. CKs encapsulated within EVs secreted by immune cells offer valuable insights into disease progression. This review presents a comprehensive analysis summarizing the relationship between CKs and SCZ and emphasizes the vital role of CKs encapsulated within EVs in the pathogenesis and development of SCZ.