Machine Learning and Non-Affective Psychosis: Identification, Differential Diagnosis, and Treatment

Abstract

Purpose of Review This review will cover the most relevant findings on the use of machine learning (ML) techniques in the field of non-affective psychosis, by summarizing the studies published in the last three years focusing on illness detection and treatment. Recent Findings Multiple ML tools that include mostly supervised approaches such as support vector machine, gradient boosting, and random forest showed promising results by applying these algorithms to various sources of data: socio-demographic information, EEG, language, digital content, blood biomarkers, neuroimaging, and electronic health records. However, the overall performance, in the binary classification case, varied from 0.49, which is to be considered very low (i.e., noise), to over 0.90. These results are fully justified by different factors, some of which may be attributable to the preprocessing of the data, the wide variety of the data, and the a-priori setting of hyperparameters. One of the main limitations of the field is the lack of stratification of results based on biological sex, given that psychosis presents differently in men and women; hence, the necessity to tailor identification tools and data analytic strategies. Summary Timely identification and appropriate treatment are key factors in reducing the consequences of psychotic disorders. In recent years, the emergence of new analytical tools based on artificial intelligence such as supervised ML approaches showed promises as a potential breakthrough in this field. However, ML applications in everyday practice are still in its infancy.

Screening for psychotic experiences and psychotic disorders in general psychiatric settings: a systematic review and meta-analysis

“…116 Besides more accessible administration formats, novel screening techniques, including digital phenotyping or machine learning algorithms based on electronic health record or speech and language data may improve screening techniques. [117][118][119] While this field remains nascent and such automated screening methods are not yet regularly available or integrated into clinical practice, "digital phenotyping" shows significant promise for improving screening while reducing clinician burden.…”

Establishment of a Public Mental Health Database for Research Purposes in the Ferrara Province: Development and Preliminary Evaluation Study

“…The main problem to overcome will be the imbalance of the data set, that is, when there is an unequal distribution of classes in the data set. In such instances, a standard machine learning technique, such as a support vector machine or random forest [ 60 , 61 ], will be applied. Moreover, each patient could potentially be considered as a distinct time series by including the temporal dimension of the treatment and by applying recurrent neural networks [ 62 , 63 ].…”

Identification of Diagnostic Schizophrenia Biomarkers Based on the Assessment of Immune and Systemic Inflammation Parameters Using Machine Learning Modeling

“…A definite level of immunoinflammatory activation, according to our data, remains also in the period of medically induced remission. However, application of ML methods has not yet been studied for verification of schizophrenia diagnosis by assessing the level of systemic inflammation markers [11,12]. At the same time, there are several investigations, in which interconnections have been established between the level of some markers of systemic inflammation and clinical characteristics of schizophrenia including the intensity of cognitive function impairment (TNF-α, IL-2, IL-6, IL-8, cortisol), neurocognitive defect (IL-1β, sIL-1RA, TNF-α), acute or chronic course of the disease (TNF-α, IL-2, IL-6, IL-8, cortisol) [22,23].…”

“…Nevertheless, there are a few works devoted to the study of some immunity parameters for objectifying the diagnosis of schizophrenia with ML methods [10]. Besides, the ML methods have not yet been applied to the datasets fully reflecting systemic characteristics of immune status (parameters of adaptive immunity, the level of inflammatory markers, content of major cytokines) [11][12][13]. Taking into consideration a complex character of immune system disorders in schizophrenia, inclusion into ML a broad panel of immunological data is a promising task for improving classification accuracy and selecting the variable parameters typical for the majority of patients.…”