Machine Learning Approaches for Clinical Psychology and Psychiatry

Editorial: Can We Accurately Screen for Attention-Deficit/Hyperactivity Disorder? Moving to a Dimensional, Multistep Process to Support Youth Development

Mental health concerns are a major source of health-related burden worldwide, including shortened life spans as a result of accidental injury, suicide, and secondary health complications.¹ Attention-deficit/hyperactivity disorder (ADHD) is emblematic in that it can be a serious concern in and of itself as well as an early-emerging risk factor for a variety of other serious physical and mental health outcomes.² While there remains a need for more effective treatment, especially for core ADHD symptoms,³ many of the long-term negative outcomes associated with ADHD can be at least partially mitigated with early intervention.^4,5 Thus, cost-effective, accurate screening is a critical need for the field of child and adolescent psychiatry. The systematic review and meta-analysis of available ADHD screening tools by Mulraney et al.⁶ reflects an enormous effort to summarize the overall accuracy, sensitivity, and specificity of available screening measures. The results highlight both what we can and, critically, what we cannot yet achieve with current tools and suggest paths forward.

Training a machine learning classifier to identify ADHD based on real-world clinical data from medical records

The diagnostic process of attention deficit hyperactivity disorder (ADHD) is complex and relies on criteria sensitive to subjective biases. This may cause significant delays in appropriate treatment initiation. An automated analysis relying on subjective and objective measures might not only simplify the diagnostic process and reduce the time to diagnosis, but also improve reproducibility. While recent machine learning studies have succeeded at distinguishing ADHD from healthy controls, the clinical process requires differentiating among other or multiple psychiatric conditions. We trained a linear support vector machine (SVM) classifier to detect participants with ADHD in a population showing a broad spectrum of psychiatric conditions using anonymized data from clinical records (N = 299 participants). We differentiated children and adolescents with ADHD from those not having the condition with an accuracy of 66.1%. SVM using single features showed slight differences between features and overlapping standard deviations of the achieved accuracies. An automated feature selection achieved the best performance using a combination 19 features. Real-world clinical data from medical records can be used to automatically identify individuals with ADHD among help-seeking individuals using machine learning. The relevant diagnostic information can be reduced using an automated feature selection without loss of performance. A broad combination of symptoms across different domains, rather than specific domains, seems to indicate an ADHD diagnosis.

Neural Networks for Early Diagnosis of Postpartum PTSD in Women after Cesarean Section

The correlation between the kind of cesarean section and post-traumatic stress disorder (PTSD) in Greek women after a traumatic birth experience has been recognized in previous studies along with other risk factors, such as perinatal conditions and traumatic life events. Data from early studies have suggested some possible links between some vulnerable factors and the potential development of postpartum PTSD. The classification of each case in three possible states (PTSD, profile PTSD, and free of symptoms) is typically performed using the guidelines and the metrics of the version V of the Diagnostic and Statistical Manual of Mental Disorders (DSM-V) which requires the completion of several questionnaires during the postpartum period. The motivation in the present work is the need for a model that can detect possible PTSD cases using a minimum amount of information and produce an early diagnosis. The early PTSD diagnosis is critical since it allows the medical personnel to take the proper measures as soon as possible. Our sample consists of 469 women who underwent emergent or elective cesarean delivery in a university hospital in Greece. The methodology which is followed is the application of random decision forests (RDF) to detect the most suitable and easily accessible information which is then used by an artificial neural network (ANN) for the classification. As is demonstrated from the results, the derived decision model can reach high levels of accuracy even when only partial and quickly available information is provided.

Machine learning in the prediction of postpartum depression: A review

BACKGROUND

Current screening options in the setting of postpartum depression (PPD) are firmly rooted in self-report symptom-based tools. The implementation of the modern machine learning (ML) approaches might, in this context, represent a way to refine patient screening by precisely identifying possible PPD predictors and, subsequently, a population at risk of developing the disease, in an effort to lower its morbidity, mortality and its economic burden.

METHODS

We performed a bibliographic search on PubMed and Embase looking for studies aimed at the identification of PPD predictors using ML techniques.

RESULTS

Among the 482 articles retrieved, 11 met the inclusion criteria. The most used algorithm was the support vector machine. Notably, all studies reached an area under the curve above 0.7, ultimately suggesting that the prediction of PPD could be feasible. Variables obtained from sociodemographic and clinical aspects (psychiatric and gynecological factors) seem to be the most reliable. Only three studies employed biological variables, in the form of blood, genetic and epigenetic predictors, while no study employed imaging techniques.

LIMITATIONS

The literature on PPD prediction via ML techniques is currently scarce, with most studies employing different variables selection and ML algorithms, ultimately reducing the generalizability of the results.

CONCLUSIONS

The identification of a population at risk of developing PPD might be feasible with current technology and clinical knowledge. Further studies are necessary to clarify how such an approach could be implemented into clinical practice.

Creating sparser prediction models of treatment outcome in depression: a proof-of-concept study using simultaneous feature selection and hyperparameter tuning

Background

Predicting treatment outcome in major depressive disorder (MDD) remains an essential challenge for precision psychiatry. Clinical prediction models (CPMs) based on supervised machine learning have been a promising approach for this endeavor. However, only few CPMs have focused on model sparsity even though sparser models might facilitate the translation into clinical practice and lower the expenses of their application.

Methods

In this study, we developed a predictive modeling pipeline that combines hyperparameter tuning and recursive feature elimination in a nested cross-validation framework. We applied this pipeline to a real-world clinical data set on MDD treatment response and to a second simulated data set using three different classification algorithms. Performance was evaluated by permutation testing and comparison to a reference pipeline without nested feature selection.

Results

Across all models, the proposed pipeline led to sparser CPMs compared to the reference pipeline. Except for one comparison, the proposed pipeline resulted in equally or more accurate predictions. For MDD treatment response, balanced accuracy scores ranged between 61 and 71% when models were applied to hold-out validation data.

Conclusions

The resulting models might be particularly interesting for clinical applications as they could reduce expenses for clinical institutions and stress for patients.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12911-022-01926-2.

Cognition Meets Gait: Where and How Mind and Body Weave Each Other in a Computational Psychometrics Approach in Aging

Aging may be associated with conditions characterized by motor and cognitive alterations, which could have a detrimental impact on daily life. Although motors and cognitive aspects have always been treated as separate entities, recent literature highlights their relationship, stressing a strong association between locomotion and executive functions. Thus, designing interventions targeting the risks deriving from both components’ impairments is crucial: the dual-task represents a starting point. Although its role in targeting and decreasing difficulties in aging is well known, most interventions are focused on a single domain, proposing a vertical model in which patients emerge only for a single aspect per time during assessment and rehabilitation. In this perspective, we propose a view of the individual as a whole between mind and body, suggesting a multicomponent and multidomain approach that could integrate different domains at the same time retracing lifelike situations. Virtual Reality, thanks to the possibility to develop daily environments with engaging challenges for patients, as well as to manage different devices to collect multiple data, provides the optimal scenario in which the integration could occur. Artificial Intelligence, otherwise, offers the best methodologies to integrate a great amount of various data to create a predictive model and identify appropriate and individualized interventions. Based on these assumptions the present perspective aims to propose the development of a new approach to an integrated, multimethod, multidimensional training in order to enhance cognition and physical aspects based on behavioral data, incorporating consolidated technologies in an innovative approach to neurology.

Comprehensive Analysis of LINC01615 in Head and Neck Squamous Cell Carcinoma: A Hub Biomarker Identified by Machine Learning and Experimental Validation

Background

Head and neck squamous cell carcinoma (HNSCC) is one of the most common cancers, but in clinical practice, the lack of precise biomarkers often results in an advanced diagnosis. Hence, it is crucial to explore novel biomarkers to improve the clinical outcome of HNSCC patients.

Methods

We downloaded RNA-seq data consisting of 502 HNSCC tissues and 44 normal tissues from the TCGA database, and lncRNA genomic sequence information was downloaded from the GENECODE database for annotating lncRNA expression profiles. We used Cox regression analysis to screen prognostic lncRNAs, the threshold as HR >1 and p value <0.05. Subsequently, three survival outcomes (overall survival, progress-free interval, and disease-specific survival)-related lncRNAs overlapped to get the common lncRNAs. The hub biomarker was identified using LASSO and random forest models. Subsequently, we used a variety of statistical methods to validate the prognostic ability of the hub marker. In addition, Spearman correlation analysis between the hub marker expression and genomic heterogeneity was conducted, such as instability (MSI), homologous recombination deficiency (HRD), and tumor mutational burden (TMB). Finally, we used enrichment analysis, ssGSEA, and ESTIMATE algorithms to explore the changes in the underlying immune-related pathway and function. Finally, the MTT assay and transwell assay were performed to determine the effect of LINC01615 silencing on tumor cell proliferation, invasion, and migration.

Results

Cox regression analysis revealed 133 lncRNAs with multiple prognostic significance. The machine learning algorithm screened out the hub lncRNA with the highest importance in the RF model: LINC01615. Clinical correlation analysis revealed that the LINC01615 increased with increasing the T stage, N stage, pathology grade, and clinical stage. LINC01615 could be used as a predictor of HNSCC prognosis validating by a variety of statistical methods. Subsequently, when clinical indicators were combined with the LINC01615 expression, the visualization model (nomogram) was more applicable to clinical practice. Finally, immune algorithms indicated that LINC01615 may be involved in the regulation of lymphocyte recruitment and immunological infiltration in HNSCC, and the LINC01615 expression represented genomic heterogeneity in pan-cancer. Functionally, silencing of LINC01615 suppresses cell proliferation, invasion, and migration in HEP-2 and TU212 cells.

Conclusion

LINC01615 may play an important role in the prostromal cell enrichment and immunosuppressive state and serve as a prognostic biomarker in HNSCC.

Machine learning based prediction and the influence of complement - Coagulation pathway proteins on clinical outcome: Results from the NEURAPRO trial

BACKGROUND

Functional outcomes are important measures in the overall clinical course of psychosis and individuals at clinical high-risk (CHR), however, prediction of functional outcome remains difficult based on clinical information alone. In the first part of this study, we evaluated whether a combination of biological and clinical variables could predict future functional outcome in CHR individuals. The complement and coagulation pathways have previously been identified as being of relevance to the pathophysiology of psychosis and have been found to contribute to the prediction of clinical outcome in CHR participants. Hence, in the second part we extended the analysis to evaluate specifically the relationship of complement and coagulation proteins with psychotic symptoms and functional outcome in CHR.

MATERIALS AND METHODS

We carried out plasma proteomics and measured plasma cytokine levels, and erythrocyte membrane fatty acid levels in a sub-sample (n = 158) from the NEURAPRO clinical trial at baseline and 6 months follow up. Functional outcome was measured using Social and Occupational Functional assessment Score (SOFAS) scale. Firstly, we used support vector machine learning techniques to develop predictive models for functional outcome at 12 months. Secondly, we developed linear regression models to understand the association between 6-month follow-up levels of complement and coagulation proteins with 6-month follow-up measures of positive symptoms summary (PSS) scores and functional outcome.

RESULTS AND CONCLUSION

A prediction model based on clinical and biological data including the plasma proteome, erythrocyte fatty acids and cytokines, poorly predicted functional outcome at 12 months follow-up in CHR participants. In linear regression models, four complement and coagulation proteins (coagulation protein X, Complement C1r subcomponent like protein, Complement C4A & Complement C5) indicated a significant association with functional outcome; and two proteins (coagulation factor IX and complement C5) positively associated with the PSS score. Our study does not provide support for the utility of cytokines, proteomic or fatty acid data for prediction of functional outcomes in individuals at high-risk for psychosis. However, the association of complement protein levels with clinical outcome suggests a role for the complement system and the activity of its related pathway in the functional impairment and positive symptom severity of CHR patients.

Data-driven cognitive phenotypes in subjects with bipolar disorder and their clinical markers of severity

BACKGROUND

Subjects with bipolar disorder (BD) show heterogeneous cognitive profile and that not necessarily the disease will lead to unfavorable clinical outcomes. We aimed to identify clinical markers of severity among cognitive clusters in individuals with BD through data-driven methods.

METHODS

We recruited 167 outpatients with BD and 100 unaffected volunteers from Brazil and Spain that underwent a neuropsychological assessment. Cognitive functions assessed were inhibitory control, processing speed, cognitive flexibility, verbal fluency, working memory, short- and long-term verbal memory. We performed hierarchical cluster analysis and discriminant function analysis to determine and confirm cognitive clusters, respectively. Then, we used classification and regression tree (CART) algorithm to determine clinical and sociodemographic variables of the previously defined cognitive clusters.

RESULTS

We identified three neuropsychological subgroups in individuals with BD: intact (35.3%), selectively impaired (34.7%), and severely impaired individuals (29.9%). The most important predictors of cognitive subgroups were years of education, the number of hospitalizations, and age, respectively. The model with CART algorithm showed sensitivity 45.8%, specificity 78.4%, balanced accuracy 62.1%, and the area under the ROC curve was 0.61. Of 10 attributes included in the model, only three variables were able to separate cognitive clusters in BD individuals: years of education, number of hospitalizations, and age.

CONCLUSION

These results corroborate with recent findings of neuropsychological heterogeneity in BD, and suggest an overlapping between premorbid and morbid aspects that influence distinct cognitive courses of the disease.

Lessons Learned and Future Directions of MetaTutor: Leveraging Multichannel Data to Scaffold Self-Regulated Learning With an Intelligent Tutoring System

Self-regulated learning (SRL) is critical for learning across tasks, domains, and contexts. Despite its importance, research shows that not all learners are equally skilled at accurately and dynamically monitoring and regulating their self-regulatory processes. Therefore, learning technologies, such as intelligent tutoring systems (ITSs), have been designed to measure and foster SRL. This paper presents an overview of over 10 years of research on SRL with MetaTutor, a hypermedia-based ITS designed to scaffold college students' SRL while they learn about the human circulatory system. MetaTutor's architecture and instructional features are designed based on models of SRL, empirical evidence on human and computerized tutoring principles of multimedia learning, Artificial Intelligence (AI) in educational systems for metacognition and SRL, and research on SRL from our team and that of other researchers. We present MetaTutor followed by a synthesis of key research findings on the effectiveness of various versions of the system (e.g., adaptive scaffolding vs. no scaffolding of self-regulatory behavior) on learning outcomes. First, we focus on findings from self-reports, learning outcomes, and multimodal data (e.g., log files, eye tracking, facial expressions of emotion, screen recordings) and their contributions to our understanding of SRL with an ITS. Second, we elaborate on the role of embedded pedagogical agents (PAs) as external regulators designed to scaffold learners' cognitive and metacognitive SRL strategy use. Third, we highlight and elaborate on the contributions of multimodal data in measuring and understanding the role of cognitive, affective, metacognitive, and motivational (CAMM) processes. Additionally, we unpack some of the challenges these data pose for designing real-time instructional interventions that scaffold SRL. Fourth, we present existing theoretical, methodological, and analytical challenges and briefly discuss lessons learned and open challenges.

Individualized identification of sexual dysfunction of psychiatric patients with machine-learning

Sexual dysfunction (SD) is prevalent in patients with mental health disorders and can significantly impair their quality of life. Early recognition of SD in a clinical setting may help patients and clinicians to optimize treatment options of SD and/or other primary diagnoses taking SD risk into account and may facilitate treatment compliance. SD identification is often overlooked in clinical practice; we seek to explore whether patients with a high risk of SD can be identified at the individual level by assessing known risk factors via a machine learning (ML) model. We assessed 135 subjects referred to a tertiary mental health clinic in a Western Canadian city using health records data, including age, sex, physician’s diagnoses, drug treatment, and the Arizona Sexual Experiences Scale (ASEX). A ML model was fitted to the data, with SD status derived from the ASEX as target outcomes and all other variables as predicting variables. Our ML model was able to identify individual SD cases—achieving a balanced accuracy of 0.736, with a sensitivity of 0.750 and a specificity of 0.721—and identified major depressive disorder and female sex as risk factors, and attention deficit hyperactivity disorder as a potential protective factor. This study highlights the utility of SD screening in a psychiatric clinical setting, demonstrating a proof-of-concept ML approach for SD screening in psychiatric patients, which has marked potential to improve their quality of life.

A Genomic Information Management System for Maintaining Healthy Genomic States and Application of Genomic Big Data in Clinical Research

Improvements in next-generation sequencing (NGS) technology and computer systems have enabled personalized therapies based on genomic information. Recently, health management strategies using genomics and big data have been developed for application in medicine and public health science. In this review, I first discuss the development of a genomic information management system (GIMS) to maintain a highly detailed health record and detect diseases by collecting the genomic information of one individual over time. Maintaining a health record and detecting abnormal genomic states are important; thus, the development of a GIMS is necessary. Based on the current research status, open public data, and databases, I discuss the possibility of a GIMS for clinical use. I also discuss how the analysis of genomic information as big data can be applied for clinical and research purposes. Tremendous volumes of genomic information are being generated, and the development of methods for the collection, cleansing, storing, indexing, and serving must progress under legal regulation. Genetic information is a type of personal information and is covered under privacy protection; here, I examine the regulations on the use of genetic information in different countries. This review provides useful insights for scientists and clinicians who wish to use genomic information for healthy aging and personalized medicine.

Neuroanatomical markers of psychotic experiences in adolescents: A machine-learning approach in a longitudinal population-based sample

It is important to identify accurate markers of psychiatric illness to aid early prediction of disease course. Subclinical psychotic experiences (PEs) are important risk factors for later mental ill-health and suicidal behaviour. This study used machine learning to investigate neuroanatomical markers of PEs in early and later stages of adolescence. Machine learning using logistic regression using Elastic Net regularization was applied to T1-weighted and diffusion MRI data to classify adolescents with subclinical psychotic experiences vs. controls across 3 timepoints (Time 1:11-13 years, n = 77; Time 2:14-16 years, n = 56; Time 3:18-20 years, n = 40). Neuroimaging data classified adolescents aged 11-13 years with current PEs vs. controls returning an AROC of 0.62, significantly better than a null model, p = 1.73^e-29. Neuroimaging data also classified those with PEs at 18-20 years (AROC = 0.59;P = 7.19^e-10) but performance was at chance level at 14-16 years (AROC = 0.50). Left hemisphere frontal regions were top discriminant classifiers for 11-13 years-old adolescents with PEs, particularly pars opercularis. Those with future PEs at 18-20 years-old were best distinguished from controls based on left frontal regions, right-hemisphere medial lemniscus, cingulum bundle, precuneus and genu of the corpus callosum (CC). Deviations from normal adolescent brain development in young people with PEs included an acceleration in the typical pattern of reduction in left frontal thickness and right parietal curvature, and accelerated progression of microstructural changes in right white matter and corpus callosum. These results emphasise the importance of multi-modal analysis for understanding adolescent PEs and provide important new insights into early phenotypes for psychotic experiences.

A machine learning approach to predict resilience and sickness absence in the healthcare workforce during the COVID-19 pandemic

During the COVID-19 pandemic, healthcare workers (HCWs) have faced unprecedented workloads and personal health risks leading to mental disorders and surges in sickness absence. Previous work has shown that interindividual differences in psychological resilience might explain why only some individuals are vulnerable to these consequences. However, no prognostic tools to predict individual HCW resilience during the pandemic have been developed. We deployed machine learning (ML) to predict psychological resilience during the pandemic. The models were trained in HCWs of the largest Finnish hospital, Helsinki University Hospital (HUS, N = 487), with a six-month follow-up, and prognostic generalizability was evaluated in two independent HCW validation samples (Social and Health Services in Kymenlaakso: Kymsote, N = 77 and the City of Helsinki, N = 322) with similar follow-ups never used for training the models. Using the most predictive items to predict future psychological resilience resulted in a balanced accuracy (BAC) of 72.7–74.3% in the HUS sample. Similar performances (BAC = 67–77%) were observed in the two independent validation samples. The models' predictions translated to a high probability of sickness absence during the pandemic. Our results provide the first evidence that ML techniques could be harnessed for the early detection of COVID-19-related distress among HCWs, thereby providing an avenue for potential targeted interventions.

Potential benefits and limitations of machine learning in the field of eating disorders: current research and future directions

Advances in machine learning and digital data provide vast potential for mental health predictions. However, research using machine learning in the field of eating disorders is just beginning to emerge. This paper provides a narrative review of existing research and explores potential benefits, limitations, and ethical considerations of using machine learning to aid in the detection, prevention, and treatment of eating disorders. Current research primarily uses machine learning to predict eating disorder status from females' responses to validated surveys, social media posts, or neuroimaging data often with relatively high levels of accuracy. This early work provides evidence for the potential of machine learning to improve current eating disorder screening methods. However, the ability of these algorithms to generalise to other samples or be used on a mass scale is only beginning to be explored. One key benefit of machine learning over traditional statistical methods is the ability of machine learning to simultaneously examine large numbers (100s to 1000s) of multimodal predictors and their complex non-linear interactions, but few studies have explored this potential in the field of eating disorders. Machine learning is also being used to develop chatbots to provide psychoeducation and coping skills training around body image and eating disorders, with implications for early intervention. The use of machine learning to personalise treatment options, provide ecological momentary interventions, and aid the work of clinicians is also discussed. Machine learning provides vast potential for the accurate, rapid, and cost-effective detection, prevention, and treatment of eating disorders. More research is needed with large samples of diverse participants to ensure that machine learning models are accurate, unbiased, and generalisable to all people with eating disorders. There are important limitations and ethical considerations with utilising machine learning methods in practice. Thus, rather than a magical solution, machine learning should be seen as an important tool to aid the work of researchers, and eventually clinicians, in the early identification, prevention, and treatment of eating disorders.

Machine learning models to predict neuropsychiatric disorders in various brain tumors

OBJECTIVE

Neuropsychiatric disorders in brain tumor patients are commonly observed. It is difficult to anticipate these disorders in different types of brain tumors. The goal of the study was to see how well machine learning (ML)-based decision algorithms might predict neuropsychiatric problems in different types of brain tumors.

METHODS

145 histopathologically-confirmed primary brain tumors of both gender aged 25-65 years of age, were included for neuropsychiatric assessments. The datasets of brain tumor patients were employed for building the models. Four different decision ML classification trees/models (J48, Random Forest, Random Tree & Hoeffding Tree) with supervised learning were trained, tested, and validated on class labeled data of brain tumor patients. The models were compared in order to determine the best accurate classifier in predicting neuropsychiatric problems in various brain tumors. Following categorical attributes as independent variables (predictors) were included from the data of brain tumor patients: age, gender, depression, dementia, and brain tumor types. With the machine learning decision tree/model techniques, a multi-target classification was performed with classes of neuropsychiatric diseases that were predicted from the selected attributes.

RESULTS

86 percent of patients were depressed, and 55 percent were suffering from dementia. Anger was the most often reported neuropsychiatric condition in brain tumor patients (92.41%), followed by sleep disorders (83%), apathy (80%), and mood swings (76.55%). When compared to other tumor types, glioblastoma patients had a higher rate of depression (20%) and dementia (20.25%). The developed models Random Forest and Random Tree were found successful with an accuracy of up to 94% (10-folds) for the prediction of neuropsychiatric disorders in brain tumor patients. The multiclass target (neuropsychiatric ailments) accuracies were having good measures of precision (0.9-1.0), recall (0.9-1.0), F-measure (0.9-1.0), and ROC area (0.9-1.0) in decision models.

CONCLUSION

Random Forest Trees can be used to accurately predict neuropsychiatric illnesses. Based on the model output, the ML-decision trees will aid the physician in pre-diagnosing the mental issue and deciding on the best therapeutic approach to avoid subsequent neuropsychiatric issues in brain tumor patients.

Identifying posttraumatic stress disorder staging from clinical and sociodemographic features: a proof-of-concept study using a machine learning approach

This proof-of-concept study aimed to investigate the viability of a predictive model to support posttraumatic stress disorder (PTSD) staging. We performed a naturalistic, cross-sectional study at two Brazilian centers: the Psychological Trauma Research and Treatment (NET-Trauma) Program at Universidade Federal of Rio Grande do Sul, and the Program for Research and Care on Violence and PTSD (PROVE), at Universidade Federal of São Paulo. Five supervised machine-learning algorithms were tested: Elastic Net, Gradient Boosting Machine, Random Forest, Support Vector Machine, and C5.0, using clinical (Clinician-Administered PTSD Scale version 5) and sociodemographic features. A hundred and twelve patients were enrolled (61 from NET-Trauma and 51 from PROVE). We found a model with four classes suitable for the PTSD staging, with best performance metrics using the C5.0 algorithm to CAPS-5 15-items plus sociodemographic features, with an accuracy of 65.6% for the train dataset and 52.9% for the test dataset (both significant). The number of symptoms, CAPS-5 total score, global severity score, and presence of current/previous trauma events appear as main features to predict PTSD staging. This is the first study to evaluate staging in PTSD with machine learning algorithms using accessible clinical and sociodemographic features, which may be used in future research.

Predictive Modeling of Mental Illness Onset Using Wearable Devices and Medical Examination Data: Machine Learning Approach

The prevention and treatment of mental illness is a serious social issue. Prediction and intervention, however, have been difficult because of lack of objective biomarkers for mental illness. The objective of this study was to use biometric data acquired from wearable devices as well as medical examination data to build a predictive model that can contribute to the prevention of the onset of mental illness. This was an observational study of 4,612 subjects from the health database of society-managed health insurance in Japan provided by JMDC Inc. The inputs to the predictive model were 3-months of continuous wearable data and medical examinations within and near that period; the output was the presence or absence of mental illness over the following month, as defined by insurance claims data. The features relating to the wearable data were sleep, activity, and resting heart rate, measured by a consumer-grade wearable device (specifically, Fitbit). The predictive model was built using the XGBoost algorithm and presented an area-under-the-receiver-operating-characteristic curve of 0.712 (SD = 0.02, a repeated stratified group 10-fold cross validation). The top-ranking feature importance measure was wearable data, and its importance was higher than the blood-test values from medical examinations. Detailed verification of the model showed that predictions were made based on disrupted sleep rhythms, mild physical activity duration, alcohol use, and medical examination data on disrupted eating habits as risk factors. In summary, the predictive model showed useful accuracy for grouping the risk of mental illness onset, suggesting the potential of predictive detection, and preventive intervention using wearable devices. Sleep abnormalities in particular were detected as wearable data 3 months prior to mental illness onset, and the possibility of early intervention targeting the stabilization of sleep as an effective measure for mental illness onset was shown.

Predicting the Severity of Lockdown-Induced Psychiatric Symptoms with Machine Learning

During the COVID-19 pandemic, an increase in the incidence of psychiatric disorders in the general population and an increase in the severity of symptoms in psychiatric patients have been reported. Anxiety and depression symptoms are the most commonly observed during large-scale dramatic events such as pandemics and wars, especially when these implicate an extended lockdown. The early detection of higher risk clinical and non-clinical individuals would help prevent the new onset and/or deterioration of these symptoms. This in turn would lead to the implementation of public policies aimed at protecting vulnerable populations during these dramatic contingencies, therefore optimising the effectiveness of interventions and saving the resources of national healthcare systems. We used a supervised machine learning method to identify the predictors of the severity of psychiatric symptoms during the Italian lockdown due to the COVID-19 pandemic. Via a case study, we applied this methodology to a small sample of healthy individuals, obsessive-compulsive disorder patients, and adjustment disorder patients. Our preliminary results show that our models were able to predict depression, anxiety, and obsessive-compulsive symptoms during the lockdown with up to 92% accuracy based on demographic and clinical characteristics collected before the pandemic. The presented methodology may be used to predict the psychiatric prognosis of individuals under a large-scale lockdown and thus supporting the related clinical decisions.

Predicting How Well Adolescents Get Along with Peers and Teachers: A Machine Learning Approach

How well adolescents get along with others such as peers and teachers is an important aspect of adolescent development. Current research on adolescent relationship with peers and teachers is limited by classical methods that lack explicit test of predictive performance and cannot efficiently discover complex associations with potential non-linearity and higher-order interactions among a large set of predictors. Here, a transparently reported machine learning approach is utilized to overcome these limitations in concurrently predicting how well adolescents perceive themselves to get along with peers and teachers. The predictors were 99 items from four instruments examining internalizing and externalizing psychopathology, sensation-seeking, peer pressure, and parent-child conflict. The sample consisted of 3232 adolescents (M = 14.0 years, SD = 1.0 year, 49% female). Nonlinear machine learning classifiers predicted with high performance adolescent relationship with peers and teachers unlike classical methods. Using model explainability analyses at the item level, results identified influential predictors related to somatic complaints and attention problems that interacted in nonlinear ways with internalizing behaviors. In many cases, these intrapersonal predictors outcompeted in predictive power many interpersonal predictors. Overall, the results suggest the need to cast a much wider net of variables for understanding and predicting adolescent relationships, and highlight the power of a data-driven machine learning approach with implications on a predictive science of adolescence research.

Praxisorientierte Forschung in der Psychotherapie

Zusammenfassung. In den letzten Jahrzehnten hat sich durch randomisiert-kontrollierte Studien (RCTs) eine breite Evidenzbasis von Psychotherapie mit mittleren bis großen Effekten für verschiedene psychische Störungen gebildet. Neben der Bestimmung dieser Wirksamkeit („Efficacy“) ebneten Studien zur Wirksamkeit unter alltäglichen Routinebedingungen („Effectiveness“) historisch den Weg zur Entwicklung eines praxisorientierten Forschungsparadigmas. Im Beitrag wird argumentiert, dass im Rahmen dieses Paradigmas praxisbasierte Studien eine wertvolle Ergänzung zu RCTs darstellen, da sie existierende Probleme in der Psychotherapieforschung adressieren können. In der gegenwärtigen praxisorientierten Forschung liefern dabei neue Ansätze aus der personalisierten Medizin und Methoden aus der ‚Computational Psychiatry‘ wichtige Anhaltspunkte zur Optimierung von Effekten in der Psychotherapie. Im Kontext der Personalisierung werden bspw. klinische multivariable Prädiktionsmodelle entwickelt, welche durch Rückmeldeschleifen an Praktiker_innen kurzfristig ein evidenzbasiertes Outcome-Monitoring ermöglicht und langfristig das Praxis-Forschungsnetzwerk in Deutschland stärkt. Am Ende des Beitrags werden zukünftige Richtungen für die praxisorientierte Forschung im Sinne des ‘Precision Mental Health Care’ -Paradigmas abgeleitet und diskutiert.

Five points to consider when reading a translational machine-learning paper

Machine-learning techniques are used in this BJPsych special issue on precision medicine in attempts to create statistical models that make clinically relevant predictions for individual patients. In this primer, we outline five key points that are helpful for a new reader to consider in order to engage with the field and evaluate the literature. These points include the consideration of why we are interested in new statistical approaches, how they may produce individualised predictions, what caveats need to be kept in-mind and why the interest and engagment of clinicians and clinical researchers is critical to successful model development and implementation. We hope that the following primer will provide shared understanding to encourage dialogue between clinical and methodological fields.

Using machine learning to predict heavy drinking during outpatient alcohol treatment

BACKGROUND

Accurate clinical prediction supports the effective treatment of alcohol use disorder (AUD) and other psychiatric disorders. Traditional statistical techniques have identified patient characteristics associated with treatment outcomes. However, less work has focused on systematically leveraging these associations to create optimal predictive models. The current study demonstrates how machine learning can be used to predict clinical outcomes in people completing outpatient AUD treatment.

METHOD

We used data from the COMBINE multisite clinical trial (n = 1383) to develop and test predictive models. We identified three priority prediction targets, including (1) heavy drinking during the first month of treatment, (2) heavy drinking during the last month of treatment, and (3) heavy drinking between weekly/bi-weekly sessions. Models were generated using the random forest algorithm. We used "leave sites out" partitioning to externally validate the models in trial sites that were not included in the model training. Stratified model development was used to test for sex differences in the relative importance of predictive features.

RESULTS

Models predicting heavy alcohol use during the first and last months of treatment showed internal cross-validation area under the curve (AUC) scores ranging from 0.67 to 0.74. AUC was comparable in the external validation using data from held-out sites (AUC range = 0.69 to 0.72). The model predicting between-session heavy drinking showed strong classification accuracy in internal cross-validation (AUC = 0.89) and external test samples (AUC range = 0.80 to 0.87). Stratified analyses showed substantial sex differences in optimal feature sets.

CONCLUSION

Machine learning techniques can predict alcohol treatment outcomes using routinely collected clinical data. This technique has the potential to greatly improve clinical prediction accuracy without requiring expensive or invasive assessment methods. More research is needed to understand how best to deploy these models.

Annual Research Review: Translational machine learning for child and adolescent psychiatry

Children and adolescents could benefit from the use of predictive tools that facilitate personalized diagnoses, prognoses, and treatment selection. Such tools have not yet been deployed using traditional statistical methods, potentially due to the limitations of the paradigm and the need to leverage large amounts of digital data. This review will suggest that a machine learning approach could address these challenges and is designed to introduce new readers to the background, methods, and results in the field. A rationale is first introduced followed by an outline of fundamental elements of machine learning approaches. To provide an overview of the use of the techniques in child and adolescent literature, a scoping review of broad trends is then presented. Selected studies are also highlighted in order to draw attention to research areas that are closest to translation and studies that exhibit a high degree of experimental innovation. Limitations to the research, and machine learning approaches generally, are outlined in the penultimate section highlighting issues related to sample sizes, validation, clinical utility, and ethical challenges. Finally, future directions are discussed that could enhance the possibility of clinical implementation and address specific questions relevant to the child and adolescent psychiatry. The review gives a broad overview of the machine learning paradigm in order to highlight the benefits of a shift in perspective towards practically oriented statistical solutions that aim to improve clinical care of children and adolescents.

Machine Learning Algorithms for Depression: Diagnosis, Insights, and Research Directions

Over the years, stress, anxiety, and modern-day fast-paced lifestyles have had immense psychological effects on people’s minds worldwide. The global technological development in healthcare digitizes the scopious data, enabling the map of the various forms of human biology more accurately than traditional measuring techniques. Machine learning (ML) has been accredited as an efficient approach for analyzing the massive amount of data in the healthcare domain. ML methodologies are being utilized in mental health to predict the probabilities of mental disorders and, therefore, execute potential treatment outcomes. This review paper enlists different machine learning algorithms used to detect and diagnose depression. The ML-based depression detection algorithms are categorized into three classes, classification, deep learning, and ensemble. A general model for depression diagnosis involving data extraction, pre-processing, training ML classifier, detection classification, and performance evaluation is presented. Moreover, it presents an overview to identify the objectives and limitations of different research studies presented in the domain of depression detection. Furthermore, it discussed future research possibilities in the field of depression diagnosis.

Prediction of quality of life in schizophrenia using machine learning models on data from Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) schizophrenia trial

While research focus remains mainly on psychotic symptoms, it is questionable whether we are placing enough emphasis on improving the quality of life (QoL) of schizophrenia patients. To date, the predictive power of QoL remained limited. Therefore, this study aimed to accurately predict the QoL within schizophrenia using supervised learning methods. The authors report findings from participants of a large randomized, double-blind clinical trial for schizophrenia treatment. Potential predictors of QoL included all available and non-redundant variables from the dataset. By optimizing parameters, three linear LASSO regressions were calculated (N = 697, 692, and 786), including 44, 47, and 41 variables, with adjusted R-squares ranging from 0.31 to 0.36. Best predictors included social and emotion-related symptoms, neurocognition (processing speed), education, female gender, treatment attitudes, and mental, emotional, and physical health. These results demonstrate that machine learning is an excellent predictive tool to process clinical data. It appears that the patient’s perception of their treatment has an important impact on patients’ QoL and that interventions should consider this aspect.

Trial registration: ClinicalTrials.gov Identifier: NCT00014001.

Digital phenotyping in depression diagnostics: Integrating psychiatric and engineering perspectives

Depression is a serious medical condition and is a leading cause of disability worldwide. Current depression diagnostics and assessment has significant limitations due to heterogeneity of clinical presentations, lack of objective assessments, and assessments that rely on patients' perceptions, memory, and recall. Digital phenotyping (DP), especially assessments conducted using mobile health technologies, has the potential to greatly improve accuracy of depression diagnostics by generating objectively measurable endophenotypes. DP includes two primary sources of digital data generated using ecological momentary assessments (EMA), assessments conducted in real-time, in subjects' natural environment. This includes active EMA, data that require active input by the subject, and passive EMA or passive sensing, data passively and automatically collected from subjects' personal digital devices. The raw data is then analyzed using machine learning algorithms to identify behavioral patterns that correlate with patients' clinical status. Preliminary investigations have also shown that linguistic and behavioral clues from social media data and data extracted from the electronic medical records can be used to predict depression status. These other sources of data and recent advances in telepsychiatry can further enhance DP of the depressed patients. Success of DP endeavors depends on critical contributions from both psychiatric and engineering disciplines. The current review integrates important perspectives from both disciplines and discusses parameters for successful interdisciplinary collaborations. A clinically-relevant model for incorporating DP in clinical setting is presented. This model, based on investigations conducted by our group, delineates development of a depression prediction system and its integration in clinical setting to enhance depression diagnostics and inform the clinical decision making process. Benefits, challenges, and opportunities pertaining to clinical integration of DP of depression diagnostics are discussed from interdisciplinary perspectives.

Pattern of predictive features of continued cannabis use in patients with recent-onset psychosis and clinical high-risk for psychosis

Continued cannabis use (CCu) is an important predictor for poor long-term outcomes in psychosis and clinically high-risk patients, but no generalizable model has hitherto been tested for its ability to predict CCu in these vulnerable patient groups. In the current study, we investigated how structured clinical and cognitive assessments and structural magnetic resonance imaging (sMRI) contributed to the prediction of CCu in a group of 109 patients with recent-onset psychosis (ROP). We tested the generalizability of our predictors in 73 patients at clinical high-risk for psychosis (CHR). Here, CCu was defined as any cannabis consumption between baseline and 9-month follow-up, as assessed in structured interviews. All patients reported lifetime cannabis use at baseline. Data from clinical assessment alone correctly classified 73% (p < 0.001) of ROP and 59 % of CHR patients. The classifications of CCu based on sMRI and cognition were non-significant (ps > 0.093), and their addition to the interview-based predictor via stacking did not improve prediction significantly, either in the ROP or CHR groups (ps > 0.065). Lower functioning, specific substance use patterns, urbanicity and a lack of other coping strategies contributed reliably to the prediction of CCu and might thus represent important factors for guiding preventative efforts. Our results suggest that it may be possible to identify by clinical measures those psychosis-spectrum patients at high risk for CCu, potentially allowing to improve clinical care through targeted interventions. However, our model needs further testing in larger samples including more diverse clinical populations before being transferred into clinical practice.

Predictors of suicide attempt in patients with obsessive-compulsive disorder: an exploratory study with machine learning analysis

BACKGROUND

Patients with obsessive-compulsive disorder (OCD) are at increased risk for suicide attempt (SA) compared to the general population. However, the significant risk factors for SA in this population remains unclear - whether these factors are associated with the disorder itself or related to extrinsic factors, such as comorbidities and sociodemographic variables. This study aimed to identify predictors of SA in OCD patients using a machine learning algorithm.

METHODS

A total of 959 outpatients with OCD were included. An elastic net model was performed to recognize the predictors of SA among OCD patients, using clinical and sociodemographic variables.

RESULTS

The prevalence of SA in our sample was 10.8%. Relevant predictors of SA founded by the elastic net algorithm were the following: previous suicide planning, previous suicide thoughts, lifetime depressive episode, and intermittent explosive disorder. Our elastic net model had a good performance and found an area under the curve of 0.95.

CONCLUSIONS

This is the first study to evaluate risk factors for SA among OCD patients using machine learning algorithms. Our results demonstrate an accurate risk algorithm can be created using clinical and sociodemographic variables. All aspects of suicidal phenomena need to be carefully investigated by clinicians in every evaluation of OCD patients. Particular attention should be given to comorbidity with depressive symptoms.

Global trends of machine learning applications in psychiatric research over 30 years: A bibliometric analysis

This bibliometric analysis aimed to identify active research areas and trends in machine learning applications within the psychiatric literature. An exponential growth in the number of related publications indexed in Web of Science during the last decade was noted. Document co-citation analysis revealed 10 clusters of knowledge, which included several mental health conditions, albeit with visible structural overlap. Several influential publications in the co-citation network were identified. Keyword trends illustrated a recent shift of focus from "psychotic" to "neurotic" conditions. Despite a relative lack of literature from the developing world, a recent rise in publications from Asian countries was observed. DATA AVAILABILITY: Bibliographic data for this study were downloaded from the Web of Science. The search strategy is included in the Supplementary file.

Translating promise into practice: a review of machine learning in suicide research and prevention

In ever more pressured health-care systems, technological solutions offering scalability of care and better resource targeting are appealing. Research on machine learning as a technique for identifying individuals at risk of suicidal ideation, suicide attempts, and death has grown rapidly. This research often places great emphasis on the promise of machine learning for preventing suicide, but overlooks the practical, clinical implementation issues that might preclude delivering on such a promise. In this Review, we synthesise the broad empirical and review literature on electronic health record-based machine learning in suicide research, and focus on matters of crucial importance for implementation of machine learning in clinical practice. The challenge of preventing statistically rare outcomes is well known; progress requires tackling data quality, transparency, and ethical issues. In the future, machine learning models might be explored as methods to enable targeting of interventions to specific individuals depending upon their level of need-ie, for precision medicine. Primarily, however, the promise of machine learning for suicide prevention is limited by the scarcity of high-quality scalable interventions available to individuals identified by machine learning as being at risk of suicide.

Using combined environmental-clinical classification models to predict role functioning outcome in clinical high-risk states for psychosis and recent-onset depression

BACKGROUND

Clinical high-risk states for psychosis (CHR) are associated with functional impairments and depressive disorders. A previous PRONIA study predicted social functioning in CHR and recent-onset depression (ROD) based on structural magnetic resonance imaging (sMRI) and clinical data. However, the combination of these domains did not lead to accurate role functioning prediction, calling for the investigation of additional risk dimensions. Role functioning may be more strongly associated with environmental adverse events than social functioning.

AIMS

We aimed to predict role functioning in CHR, ROD and transdiagnostically, by adding environmental adverse events-related variables to clinical and sMRI data domains within the PRONIA sample.

METHOD

Baseline clinical, environmental and sMRI data collected in 92 CHR and 95 ROD samples were trained to predict lower versus higher follow-up role functioning, using support vector classification and mixed k-fold/leave-site-out cross-validation. We built separate predictions for each domain, created multimodal predictions and validated them in independent cohorts (74 CHR, 66 ROD).

RESULTS

Models combining clinical and environmental data predicted role outcome in discovery and replication samples of CHR (balanced accuracies: 65.4% and 67.7%, respectively), ROD (balanced accuracies: 58.9% and 62.5%, respectively), and transdiagnostically (balanced accuracies: 62.4% and 68.2%, respectively). The most reliable environmental features for role outcome prediction were adult environmental adjustment, childhood trauma in CHR and childhood environmental adjustment in ROD.

CONCLUSIONS

Findings support the hypothesis that environmental variables inform role outcome prediction, highlight the existence of both transdiagnostic and syndrome-specific predictive environmental adverse events, and emphasise the importance of implementing real-world models by measuring multiple risk dimensions.

Viability Study of Machine Learning-Based Prediction of COVID-19 Pandemic Impact in Obsessive-Compulsive Disorder Patients

Background

Machine learning modeling can provide valuable support in different areas of mental health, because it enables to make rapid predictions and therefore support the decision making, based on valuable data. However, few studies have applied this method to predict symptoms’ worsening, based on sociodemographic, contextual, and clinical data. Thus, we applied machine learning techniques to identify predictors of symptomatologic changes in a Spanish cohort of OCD patients during the initial phase of the COVID-19 pandemic.

Methods

127 OCD patients were assessed using the Yale–Brown Obsessive-Compulsive Scale (Y-BOCS) and a structured clinical interview during the COVID-19 pandemic. Machine learning models for classification (LDA and SVM) and regression (linear regression and SVR) were constructed to predict each symptom based on patient’s sociodemographic, clinical and contextual information.

Results

A Y-BOCS score prediction model was generated with 100% reliability at a score threshold of ± 6. Reliability of 100% was reached for obsessions and/or compulsions related to COVID-19. Symptoms of anxiety and depression were predicted with less reliability (correlation R of 0.58 and 0.68, respectively). The suicidal thoughts are predicted with a sensitivity of 79% and specificity of 88%. The best results are achieved by SVM and SVR.

Conclusion

Our findings reveal that sociodemographic and clinical data can be used to predict changes in OCD symptomatology. Machine learning may be valuable tool for helping clinicians to rapidly identify patients at higher risk and therefore provide optimized care, especially in future pandemics. However, further validation of these models is required to ensure greater reliability of the algorithms for clinical implementation to specific objectives of interest.

Validation and generalizability of machine learning prediction models on attrition in longitudinal studies

Attrition in longitudinal studies is a major threat to the representativeness of the data and the generalizability of the findings. Typical approaches to address systematic nonresponse are either expensive and unsatisfactory (e.g., oversampling) or rely on the unrealistic assumption of data missing at random (e.g., multiple imputation). Thus, models that effectively predict who most likely drops out in subsequent occasions might offer the opportunity to take countermeasures (e.g., incentives). With the current study, we introduce a longitudinal model validation approach and examine whether attrition in two nationally representative longitudinal panel studies can be predicted accurately. We compare the performance of a basic logistic regression model with a more flexible, data-driven machine learning algorithm—gradient boosting machines. Our results show almost no difference in accuracies for both modeling approaches, which contradicts claims of similar studies on survey attrition. Prediction models could not be generalized across surveys and were less accurate when tested at a later survey wave. We discuss the implications of these findings for survey retention, the use of complex machine learning algorithms, and give some recommendations to deal with study attrition.

Classification of amyotrophic lateral sclerosis by brain volume, connectivity, and network dynamics

Emerging studies corroborate the importance of neuroimaging biomarkers and machine learning to improve diagnostic classification of amyotrophic lateral sclerosis (ALS). While most studies focus on structural data, recent studies assessing functional connectivity between brain regions by linear methods highlight the role of brain function. These studies have yet to be combined with brain structure and nonlinear functional features. We investigate the role of linear and nonlinear functional brain features, and the benefit of combining brain structure and function for ALS classification. ALS patients (N = 97) and healthy controls (N = 59) underwent structural and functional resting state magnetic resonance imaging. Based on key hubs of resting state networks, we defined three feature sets comprising brain volume, resting state functional connectivity (rsFC), as well as (nonlinear) resting state dynamics assessed via recurrent neural networks. Unimodal and multimodal random forest classifiers were built to classify ALS. Out-of-sample prediction errors were assessed via five-fold cross-validation. Unimodal classifiers achieved a classification accuracy of 56.35-61.66%. Multimodal classifiers outperformed unimodal classifiers achieving accuracies of 62.85-66.82%. Evaluating the ranking of individual features' importance scores across all classifiers revealed that rsFC features were most dominant in classification. While univariate analyses revealed reduced rsFC in ALS patients, functional features more generally indicated deficits in information integration across resting state brain networks in ALS. The present work undermines that combining brain structure and function provides an additional benefit to diagnostic classification, as indicated by multimodal classifiers, while emphasizing the importance of capturing both linear and nonlinear functional brain properties to identify discriminative biomarkers of ALS.

Acoustic and Facial Features From Clinical Interviews for Machine Learning-Based Psychiatric Diagnosis: Algorithm Development

BACKGROUND

In contrast to all other areas of medicine, psychiatry is still nearly entirely reliant on subjective assessments such as patient self-report and clinical observation. The lack of objective information on which to base clinical decisions can contribute to reduced quality of care. Behavioral health clinicians need objective and reliable patient data to support effective targeted interventions.

OBJECTIVE

We aimed to investigate whether reliable inferences-psychiatric signs, symptoms, and diagnoses-can be extracted from audiovisual patterns in recorded evaluation interviews of participants with schizophrenia spectrum disorders and bipolar disorder.

METHODS

We obtained audiovisual data from 89 participants (mean age 25.3 years; male: 48/89, 53.9%; female: 41/89, 46.1%): individuals with schizophrenia spectrum disorders (n=41), individuals with bipolar disorder (n=21), and healthy volunteers (n=27). We developed machine learning models based on acoustic and facial movement features extracted from participant interviews to predict diagnoses and detect clinician-coded neuropsychiatric symptoms, and we assessed model performance using area under the receiver operating characteristic curve (AUROC) in 5-fold cross-validation.

RESULTS

The model successfully differentiated between schizophrenia spectrum disorders and bipolar disorder (AUROC 0.73) when aggregating face and voice features. Facial action units including cheek-raising muscle (AUROC 0.64) and chin-raising muscle (AUROC 0.74) provided the strongest signal for men. Vocal features, such as energy in the frequency band 1 to 4 kHz (AUROC 0.80) and spectral harmonicity (AUROC 0.78), provided the strongest signal for women. Lip corner-pulling muscle signal discriminated between diagnoses for both men (AUROC 0.61) and women (AUROC 0.62). Several psychiatric signs and symptoms were successfully inferred: blunted affect (AUROC 0.81), avolition (AUROC 0.72), lack of vocal inflection (AUROC 0.71), asociality (AUROC 0.63), and worthlessness (AUROC 0.61).

CONCLUSIONS

This study represents advancement in efforts to capitalize on digital data to improve diagnostic assessment and supports the development of a new generation of innovative clinical tools by employing acoustic and facial data analysis.

Advantages of Machine Learning in Forensic Psychiatric Research—Uncovering the Complexities of Aggressive Behavior in Schizophrenia

Linear statistical methods may not be suited to the understanding of psychiatric phenomena such as aggression due to their complexity and multifactorial origins. Here, the application of machine learning (ML) algorithms offers the possibility of analyzing a large number of influencing factors and their interactions. This study aimed to explore inpatient aggression in offender patients with schizophrenia spectrum disorders (SSDs) using a suitable ML model on a dataset of 370 patients. With a balanced accuracy of 77.6% and an AUC of 0.87, support vector machines (SVM) outperformed all the other ML algorithms. Negative behavior toward other patients, the breaking of ward rules, the PANSS score at admission as well as poor impulse control and impulsivity emerged as the most predictive variables in distinguishing aggressive from non-aggressive patients. The present study serves as an example of the practical use of ML in forensic psychiatric research regarding the complex interplay between the factors contributing to aggressive behavior in SSD. Through its application, it could be shown that mental illness and the antisocial behavior associated with it outweighed other predictors. The fact that SSD is also highly associated with antisocial behavior emphasizes the importance of early detection and sufficient treatment.

Applying Evidence-Centered Design to Measure Psychological Resilience: The Development and Preliminary Validation of a Novel Simulation-Based Assessment Methodology

Modern technologies have enabled the development of dynamic game- and simulation-based assessments to measure psychological constructs. This has highlighted their potential for supplementing other assessment modalities, such as self-report. This study describes the development, design, and preliminary validation of a simulation-based assessment methodology to measure psychological resilience-an important construct for multiple life domains. The design was guided by theories of resilience, and principles of evidence-centered design and stealth assessment. The system analyzed log files from a simulated task to derive individual trajectories in response to stressors. Using slope analyses, these trajectories were indicative of four types of responses to stressors: thriving, recovery, surviving, and succumbing. Using Machine Learning, the trajectories were predictive of self-reported resilience (Connor-Davidson Resilience Scale) with high accuracy, supporting construct validity of the simulation-based assessment. These findings add to the growing evidence supporting the utility of gamified assessment of psychological constructs. Importantly, these findings address theoretical debates about the construct of resilience, adding to its theory, supporting the combination of the "trait" and "process" approaches to its operationalization.

Machine Learning Based Healthcare System for Investigating the Association Between Depression and Quality of Life

New technological innovations are changing the future of healthcare system. Identification of factors that are responsible for causing depression may lead to new experiments and treatments. Because depression as a disease is becoming a leading community health concern worldwide. Using machine learning techniques this article presents a complete methodological framework to process and explore the heterogenous data and to better understand the association between factors related to quality of life and depression. Subsequently, the experimental study is mainly divided into two parts. In the first part, a data consolidation process is presented. The relationship of data is formed and to uniquely identify each relation in data the concept of the Secure Hash Algorithm is adopted. Hashing is used to locate and index the actual items in the data because it is easier to process short hash values instead of longer strings. The second part proposed a model using both unsupervised and supervised machine learning techniques. The consolidation approach helped in providing a base for formulation and validation of the research hypothesis. The Self organizing map provided 08 cluster solution and the classification problems were taken from the clustered data to further validate the performance of the posterior probability multi-class Support Vector Machine. The expectations of the importance sampling resulted in factors responsible for causing depression. The proposed model was adopted to improve the classification performance, and the result showed classification accuracy of 91.16%.

Cognitive Performance and Learning Parameters Predict Response to Working Memory Training in Parkinson's Disease

BACKGROUND

Working memory (WM) training (WMT) is a popular intervention approach against cognitive decline in patients with Parkinson's disease (PD). However, heterogeneity in WM responsiveness suggests that WMT may not be equally efficient for all patients.

OBJECTIVE

The present study aims to evaluate a multivariate model to predict post-intervention verbal WM in patients with PD using a supervised machine learning approach. We test the predictive potential of novel learning parameters derived from the WMT and compare their predictiveness to other more commonly used domains including demographic, clinical, and cognitive data.

METHODS

37 patients with PD (age: 64.09±8.56, 48.6% female, 94.7% Hoehn & Yahr stage 2) participated in a 5-week WMT. Four random forest regression models including 1) cognitive variables only, 2) learning parameters only, 3) both cognitive and learning variables, and 4) the entire set of variables (with additional demographic and clinical data, 'all' model), were built to predict immediate and 3-month-follow-up WM.

RESULT

The 'all' model predicted verbal WM with the lowest root mean square error (RMSE) compared to the other models, at both immediate (RMSE = 0.184; 95% -CI=[0.184;0.185]) and 3-month follow-up (RMSE = 0.216; 95% -CI=[0.215;0.217]). Cognitive baseline parameters were among the most important predictors in the 'all' model. The model combining cognitive and learning parameters significantly outperformed the model solely based on cognitive variables.

CONCLUSION

Commonly assessed demographic, clinical, and cognitive variables provide robust prediction of response to WMT. Nonetheless, inclusion of training-inherent learning parameters further boosts precision of prediction models which in turn may augment training benefits following cognitive interventions in patients with PD.

Association between formal thought disorders, neurocognition and functioning in the early stages of psychosis: a systematic review of the last half-century studies

Recent review articles provided an extensive collection of studies covering many aspects of format thought disorders (FTD) among their epidemiology and phenomenology, their neurobiological underpinnings, genetics as well as their transdiagnostic prevalence. However, less attention has been paid to the association of FTD with neurocognitive and functioning deficits in the early stages of evolving psychosis. Therefore, this systematic review aims to investigate the state of the art regarding the association between FTD, neurocognition and functioning in the early stages of evolving psychotic disorders in adolescents and young adults, by following the PRISMA flowchart. A total of 106 studies were screened. We included 8 studies due to their reports of associations between FTD measures and functioning outcomes measured with different scales and 7 studies due to their reports of associations between FTD measures and neurocognition. In summary, the main findings of the included studies for functioning outcomes showed that FTD severity predicted poor social functioning, unemployment, relapses, re-hospitalisations, whereas the main findings of the included studies for neurocognition showed correlations between attentional deficits, executive functions and FTD, and highlighted the predictive potential of executive dysfunctions for sustained FTD. Further studies in upcoming years taking advantage of the acceleration in computational psychiatry would allow researchers to re-investigate the clinical importance of FTD and their role in the transition from at-risk to full-blown psychosis conditions. Employing automated computer-assisted diagnostic tools in the early stages of psychosis might open new avenues to develop targeted neuropsychotherapeutics specific to FTD.

The clinical relevance of formal thought disorder in the early stages of psychosis: results from the PRONIA study

BACKGROUND

Formal thought disorder (FTD) has been associated with more severe illness courses and functional deficits in patients with psychotic disorders. However, it remains unclear whether the presence of FTD characterises a specific subgroup of patients showing more prominent illness severity, neurocognitive and functional impairments. This study aimed to identify stable and generalizable FTD-subgroups of patients with recent-onset psychosis (ROP) by applying a comprehensive data-driven clustering approach and to test the validity of these subgroups by assessing associations between this FTD-related stratification, social and occupational functioning, and neurocognition.

METHODS

279 patients with ROP were recruited as part of the multi-site European PRONIA study (Personalised Prognostic Tools for Early Psychosis Management; www.pronia.eu). Five FTD-related symptoms (conceptual disorganization, poverty of content of speech, difficulty in abstract thinking, increased latency of response and poverty of speech) were assessed with Positive and Negative Symptom Scale (PANSS) and the Scale for the Assessment of Negative Symptoms (SANS).

RESULTS

The results with two patient subgroups showing different levels of FTD were the most stable and generalizable clustering solution (predicted clustering strength value = 0.86). FTD-High subgroup had lower scores in social (p_fdr < 0.001) and role (p_fdr < 0.001) functioning, as well as worse neurocognitive performance in semantic (p_fdr < 0.001) and phonological verbal fluency (p_fdr < 0.001), short-term verbal memory (p_fdr = 0.002) and abstract thinking (p_fdr = 0.010), in comparison to FTD-Low group.

CONCLUSIONS

Clustering techniques allowed us to identify patients with more pronounced FTD showing more severe deficits in functioning and neurocognition, thus suggesting that FTD may be a relevant marker of illness severity in the early psychosis pathway.

Scan Once, Analyse Many: Using Large Open-Access Neuroimaging Datasets to Understand the Brain

We are now in a time of readily available brain imaging data. Not only are researchers now sharing data more than ever before, but additionally large-scale data collecting initiatives are underway with the vision that many future researchers will use the data for secondary analyses. Here I provide an overview of available datasets and some example use cases. Example use cases include examining individual differences, more robust findings, reproducibility-both in public input data and availability as a replication sample, and methods development. I further discuss a variety of considerations associated with using existing data and the opportunities associated with large datasets. Suggestions for further readings on general neuroimaging and topic-specific discussions are also provided.

Método experimental e ensaios clínicos

O método experimental (ME) envolve um conjunto de procedimentos que podem ser utilizados por várias ciências para responder questões de pesquisa, refutar ou corroborar hipóteses. No presente trabalho, conduzimos uma revisão sistemática de artigos de revisão publicados em português que tiveram como foco explicar o ME. Utilizamos as bases de dados SciELO e Google Acadêmico, sem restringir a data de publicação dos artigos. Incluímos 64 artigos após a aplicação dos critérios de elegibilidade. Por meio da análise de conteúdo, sintetizamos informações desses artigos que favorecem a condução de estudos experimentais e a avaliação crítica de seus métodos. Adicionalmente, apontamos lacunas como a carência de artigos sobre estatística aplicada ao ME e uso de técnicas qualitativas.

A Multivoxel Pattern Analysis of Anhedonia During Fear Extinction: Implications for Safety Learning

BACKGROUND

Pavlovian learning processes are central to the etiology and treatment of anxiety disorders. Anhedonia and related perturbations in reward processes have been implicated in Pavlovian learning. Associations between anhedonia symptoms and neural indices of Pavlovian learning can inform transdiagnostic associations among depressive and anxiety disorders.

METHODS

Participants ages 18 to 19 years (67% female) completed a fear extinction (n = 254) and recall (n = 249) paradigm during functional magnetic resonance imaging. Symptom dimensions of general distress (common to anxiety and depression), fears (more specific to anxiety), and anhedonia-apprehension (more specific to depression) were evaluated. We trained whole-brain multivoxel pattern decoders for anhedonia-apprehension during extinction and extinction recall and tested the decoders' ability to predict anhedonia-apprehension in an external validation sample. Specificity analyses examined effects covarying for general distress and fears. Decoding was repeated within canonical brain networks to highlight candidate neurocircuitry underlying whole-brain effects.

RESULTS

Whole-brain decoder training succeeded during both tasks. Prediction of anhedonia-apprehension in the external validation sample was successful for extinction (R² = 0.047; r = 0.276, p = .002) but not extinction recall (R² < 0.001, r = -0.063, p = .492). The extinction decoder remained significantly associated with anhedonia-apprehension covarying for fears and general distress (t₁₂₁ = 3.209, p = .002). Exploratory results highlighted activity in the cognitive control, default mode, limbic, salience, and visual networks related to these effects.

CONCLUSIONS

Results suggest that patterns of brain activity during extinction, particularly in the cognitive control, default mode, limbic, salience, and visual networks, can be predictive of anhedonia symptoms. Future research should examine associations between anhedonia and extinction, including studies of exposure therapy or positive affect treatments among anhedonic individuals.

Utility of Machine Learning Approach with Neuropsychological Tests in Predicting Functional Impairment of Alzheimer's Disease

Background:

In assessing the levels of clinical impairment in dementia, a summary index of neuropsychological batteries has been widely used in describing the overall functional status.

Objective:

It remains unexamined how complex patterns of the test performances can be utilized to have specific predictive meaning when the machine learning approach is applied.

Methods:

In this study, the neuropsychological battery (CERAD-K) and assessment of functioning level (Clinical Dementia Rating scale and Instrumental Activities of Daily Living) were administered to 2,642 older adults with no impairment (n = 285), mild cognitive impairment (n = 1,057), and Alzheimer’s disease (n = 1,300). Predictive accuracy on functional impairment level with the linear models of the single total score or multiple subtest scores (Model 1, 2) and support vector regression with low or high complexity (Model 3, 4) were compared across different sample sizes.

Results:

The linear models (Model 1, 2) showed superior performance with relatively smaller sample size, while nonlinear models with low and high complexity (Model 3, 4) showed an improved accuracy with a larger dataset. Unlike linear models, the nonlinear models showed a gradual increase in the predictive accuracy with a larger sample size (n > 500), especially when the model training is allowed to exploit complex patterns of the dataset.

Conclusion:

Our finding suggests that nonlinear models can predict levels of functional impairment with a sufficient dataset. The summary index of the neuropsychological battery can be augmented for specific purposes, especially in estimating the functional status of dementia.