Separating depressive comorbidity from panic disorder: A combined functional magnetic resonance imaging and machine learning approach

Machine learning applications in preventive healthcare: A systematic literature review on predictive analytics of disease comorbidity from multiple perspectives

Artificial intelligence is constantly revolutionizing biomedical research and healthcare management. Disease comorbidity is a major threat to the quality of life for susceptible groups, especially middle-aged and elderly patients. The presence of multiple chronic diseases makes precision diagnosis challenging to realize and imposes a heavy burden on the healthcare system and economy. Given an enormous amount of accumulated health data, machine learning techniques show their capability in handling this puzzle. The present study conducts a review to uncover current research efforts in applying these methods to understanding comorbidity mechanisms and making clinical predictions considering these complex patterns. A descriptive metadata analysis of 791 unique publications aims to capture the overall research progression between January 2012 and June 2023. To delve into comorbidity-focused research, 61 of these scientific papers are systematically assessed. Four predictive analytics of tasks are detected: disease comorbidity data extraction, clustering, network, and risk prediction. It is observed that some machine learning-driven applications address inherent data deficiencies in healthcare datasets and provide a model interpretation that identifies significant risk factors of comorbidity development. Based on insights, both technical and practical, gained from relevant literature, this study intends to guide future interests in comorbidity research and draw conclusions about chronic disease prevention and diagnosis with managerial implications.

Altered resting-state cerebellar-cerebral functional connectivity in patients with panic disorder before and after treatment

The study aimed to investigate the functional connectivity (FC) between the cerebellum and intrinsic cerebral networks in patients with panic disorder (PD), and to observe changes in the cerebellar-cerebral FC following pharmacotherapy. Fifty-four patients with PD and 54 healthy controls (HCs) underwent clinical assessments and functional magnetic resonance imaging scans before and after a 5-week paroxetine treatment. Seed-based cerebellar-cerebral FC was compared between the PD and HC groups, as well as between patients with PD before and after treatment. Additionally, the correlations between FC and clinical features of PD were analyzed. Compared to HCs, patients with PD had altered cerebellar-cerebral FC in the default mode, affective-limbic, and sensorimotor networks. Moreover, a negative correlation between cerebellar-insula disconnection and the severity of depressive symptoms in patients with PD (Pearson correlation, r = -0.424, p = 0.001, Bonferroni corrected) was found. After treatment, most of the enhanced FCs observed in patients with PD at baseline returned to levels similar to those observed in HCs. However, the reduced FC at baseline did not significantly change after treatment. The findings suggest that patients with PD have specific deficits in resting-state cerebellar-cerebral FC and that paroxetine may improve PD by restoring the balance of cerebellar-cerebral FC. These findings emphasize the crucial involvement of cerebellar-cerebral FC in the neuropsychological mechanisms underlying PD and in the potential pharmacological mechanisms of paroxetine for treating PD.

Machine learning applied to functional magnetic resonance imaging in anxiety disorders

BACKGROUND

Brain functional abnormalities have been commonly reported in anxiety disorders, including generalized anxiety disorder, social anxiety disorder, panic disorder, agoraphobia, and specific phobias. The role of functional abnormalities in the discrimination of these disorders can be tested with machine learning (ML) techniques. Here, we aim to provide a comprehensive overview of ML studies exploring the potential discriminating role of functional brain alterations identified by functional magnetic resonance imaging (fMRI) in anxiety disorders.

METHODS

We conducted a search on PubMed, Web of Science, and Scopus of ML investigations using fMRI as features in patients with anxiety disorders. A total of 12 studies (resting-state fMRI n = 5, task-based fMRI n = 6, resting-state and task-based fMRI n=1) met our inclusion criteria.

RESULTS

Overall, the studies showed that, regardless of the classifiers, alterations in functional connectivity and aberrant neural activation involving the amygdala, anterior cingulate cortex, hippocampus, insula, orbitofrontal cortex, temporal pole, cerebellum, default mode network, dorsal attention network, sensory network, and affective network were able to discriminate patients with anxiety from controls, with accuracies spanning from 36 % to 94 %.

LIMITATIONS

The small sample size, different ML approaches and heterogeneity in the selection of regions included in the multivariate pattern analyses limit the conclusions of the present review.

CONCLUSIONS

ML methods using fMRI as features can distinguish patients with anxiety disorders from healthy controls, indicating that these techniques could be used as a helpful tool for the diagnosis and the development of more targeted treatments for these disorders.

Effect of BI 1358894 on Cholecystokinin-Tetrapeptide (CCK-4)-Induced Anxiety, Panic Symptoms, and Stress Biomarkers: A Phase I Randomized Trial in Healthy Males

INTRODUCTION

Depression, anxiety, and/or panic disorder are often comorbid and have a complex etiology mediated through the same neuronal network. Cholecystokinin-tetrapeptide (CCK-4), a synthetic analog of the endogenous neuropeptide cholecystokinin (CCK), is thought to be implicated in this network. The CCK-4 challenge model is an accepted method of investigating the pathophysiology of panic and has been shown to mediate neuronal activation via the transient receptor potential canonical (TRPC) ion channels.

OBJECTIVES

This study aimed to assess the pharmacodynamic effects of BI 1358894, a small-molecule inhibitor of TRPC ion channel members 4 and 5 (TRPC4/5), on CCK-4-induced anxiety/panic-like symptoms and evaluate circuit engagement.

METHODS

Twenty healthy male CCK-4-sensitive volunteers entered a Phase I, double blind, randomized, two-way cross-over, single dose, placebo-controlled trial. Randomization was to oral BI 1358894 100 mg in the fed state followed by oral placebo in the fed state, or vice versa. Treatments were administered 5 h prior to intravenous CCK-4 50 µg. The primary endpoint was maximum change from baseline of the Panic Symptom Scale (PSS) sum intensity score after CCK-4 injection. Further endpoints included the emotional faces visual analog score (EVAS), the Spielberger State-Trait Anxiety Inventory (STAI), plasma adrenocorticotropic hormone (ACTH), and serum cortisol values. The safety and tolerability of BI 1358894 was assessed based on a number of parameters including occurrence of adverse events (AEs). All pharmacodynamic, pharmacokinetic, and safety endpoints were analyzed using descriptive statistics.

RESULTS

Single oral doses of BI 1358894 were generally well tolerated by the healthy male volunteers included in this study. Adjusted mean maximum change from baseline in PSS sum intensity score was 24.4 % lower in volunteers treated with BI 1358894 versus placebo, while adjusted mean maximum change from baseline of EVAS was reduced by 19.2 % (BI 1358894 vs placebo). The STAI total score before CCK-4 injection was similar in both groups (placebo: 25.1; BI 1358894: 24.3). Relative to placebo, BI 1358894 reduced CCK-4-induced mean maximum plasma ACTH and serum cortisol values by 58.6 % and 27.3 %, respectively. Investigator-assessed drug-related AEs were reported for 13/20 participants (65.0 %). There were no serious or severe AEs, AEs of special interest, AEs leading to discontinuation of trial medication, or deaths.

CONCLUSIONS

Overall, BI 1358894 reduced psychological and physiological responses to CCK-4 compared with placebo, as measured by PSS, subjective EVAS and objectively measured stress biomarkers. BI 1358894 had a positive safety profile, and single oral doses were well tolerated by the healthy volunteers. This trial (NCT03904576/1402-0005) was registered on Clinicaltrials.gov on 05.04.19.

Behavioral and neurocognitive factors distinguishing post-traumatic stress comorbidity in substance use disorders

Significant trauma histories and post-traumatic stress disorder (PTSD) are common in persons with substance use disorders (SUD) and often associate with increased SUD severity and poorer response to SUD treatment. As such, this sub-population has been associated with unique risk factors and treatment needs. Understanding the distinct etiological profile of persons with co-occurring SUD and PTSD is therefore crucial for advancing our knowledge of underlying mechanisms and the development of precision treatments. To this end, we employed supervised machine learning algorithms to interrogate the responses of 160 participants with SUD on the multidimensional NIDA Phenotyping Assessment Battery. Significant PTSD symptomatology was correctly predicted in 75% of participants (sensitivity: 80%; specificity: 72.22%) using a classification-based model based on anxiety and depressive symptoms, perseverative thinking styles, and interoceptive awareness. A regression-based machine learning model also utilized similar predictors, but failed to accurately predict severity of PTSD symptoms. These data indicate that even in a population already characterized by elevated negative affect (individuals with SUD), especially severe negative affect was predictive of PTSD symptomatology. In a follow-up analysis of a subset of 102 participants who also completed neurocognitive tasks, comorbidity status was correctly predicted in 86.67% of participants (sensitivity: 91.67%; specificity: 66.67%) based on depressive symptoms and fear-related attentional bias. However, a regression-based analysis did not identify fear-related attentional bias as a splitting factor, but instead split and categorized the sample based on indices of aggression, metacognition, distress tolerance, and interoceptive awareness. These data indicate that within a population of individuals with SUD, aberrations in tolerating and regulating aversive internal experiences may also characterize those with significant trauma histories, akin to findings in persons with anxiety without SUD. The results also highlight the need for further research on PTSD-SUD comorbidity that includes additional comparison groups (i.e., persons with only PTSD), captures additional comorbid diagnoses that may influence the PTSD-SUD relationship, examines additional types of SUDs (e.g., alcohol use disorder), and differentiates between subtypes of PTSD.

Two-channel EEG based diagnosis of panic disorder and major depressive disorder using machine learning and non-linear dynamical methods

The current study aimed to investigate the possibility of rapid and accurate diagnoses of Panic disorder (PD) and Major depressive disorder (MDD) using machine learning. The support vector machine method was applied to 2-channel EEG signals from the frontal lobes (Fp1 and Fp2) of 149 participants to classify PD and MDD patients from healthy individuals using non-linear measures as features. We found significantly lower correlation dimension and Lempel-Ziv complexity in PD patients and MDD patients in the left hemisphere compared to healthy subjects at rest. Most importantly, we obtained a 90% accuracy in classifying MDD patients vs. healthy individuals, a 68% accuracy in classifying PD patients vs. controls, and a 59% classification accuracy between PD and MDD patients. In addition to demonstrating classification performance in a simplified setting, the observed differences in EEG complexity between subject groups suggest altered cortical processing present in the frontal lobes of PD patients that can be captured through non-linear measures. Overall, this study suggests that machine learning and non-linear measures using only 2-channel frontal EEGs are useful for aiding the rapid diagnosis of panic disorder and major depressive disorder.

Functional connectivity strength and topology differences in social phobia adolescents with and without ADHD comorbidity

Social phobia (SP) is associated with changes in functional connectivity strength and topology. However, reported changes have been heterogeneous due to small sample sizes, inconsistent methodologies, and comorbidities, such as attention-deficit/hyperactivity disorder (ADHD), which has a high comorbidity rate with SP. Furthermore, there are few studies looking at SP in an adolescent population, a critical period for the development of the social brain. This project focuses on functional connectivity strength and topological differences in social phobia patients with and without ADHD comorbidity. We examined resting-state functional MRI images from 158 subjects, including 36 SP participants without ADHD comorbidity, 60 SP participants with ADHD comorbidity, and 62 healthy controls, with an overall average age of 14.16. We used a data-driven approach to examine impaired functional connectivity in a whole-brain analysis and higher-order topological differences in functional brain networks. We identified changes in the cerebellum and default mode network in social phobia patients as a whole, with the presence of ADHD comorbidity affecting various subsystems of the default mode network. Social phobia functional connectivity networks resembled random graphs, and local connectivity patterns in the superior occipital gyrus were different due to ADHD comorbidity. These alterations may indicate impairments in self-related processing, imagery, mentalizing, and predictive processes. We then used these changes in a linear support vector machine to distinguish between each pair of groups and achieved prediction accuracy significantly above chance rates. Our study extends prior research by showing that functional connectivity changes exist at adolescence, which are affected by ADHD comorbidity. As such, these results offer a new perspective in examining neurobiological changes in SP patients.

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.

Contributions of diagnostic, cognitive, and somatovisceral information to the prediction of fear ratings in spider phobic and non-spider-fearful individuals

BACKGROUND

Physiological responding is a key characteristic of fear responses. Yet, it is unknown whether the time-consuming measurement of somatovisceral responses ameliorates the prediction of individual fear responses beyond the accuracy reached by the consideration of diagnostic (e.g., phobic vs. non phobic) and cognitive (e.g., risk estimation) factors, which can be more easily assessed.

METHOD

We applied a machine learning approach to data of an experiment, in which spider phobic and non-spider fearful participants (diagnostic factor) faced pictures of spiders. For each experimental trial, participants specified their personal risk of encountering the spider (cognitive factor), as well as their subjective fear (outcome variable) on quasi-continuous scales, while diverse somatovisceral responses were registered (heart rate, electrodermal activity, respiration, facial muscle activity).

RESULTS

The machine-learning analyses revealed that fear ratings were predominantly predictable by the diagnostic factor. Yet, when allowing for learning of individual patterns in the data, somatovisceral responses contributed additional information on the fear ratings, yielding a prediction accuracy of 81% explained variance. Moreover, heart rate prior to picture onset, but not heart rate reactivity increased predictive power.

LIMITATIONS

Fear was solely assessed by verbal reports, only 27 females were considered, and no generalization to other anxiety disorders is possible.

CONCLUSIONS

After training the algorithm to learn about individual-specific responding, somatovisceral patterns can be successfully exploited. Our findings further point to the possibility that the expectancy-related autonomic state throughout the experiment predisposes an individual to experience specific levels of fear, with less influence of the actual visual stimulations.

Systematic transdiagnostic review of magnetic-resonance imaging results: Depression, anxiety disorders and their co-occurrence

BACKGROUND

Major depressive disorder (MDD) and anxiety disorders (ANX) share core symptoms such as negative affect and often co-exist. Magnetic-resonance imaging (MRI) research suggests shared neuroanatomical/neurofunctional underpinnings. So far, studies considering transdiagnostic and disorder-specific neural alterations in MDD and ANX as well as the comorbid condition (COM) have not been reviewed systematically.

METHODS

Following PRISMA guidelines, the literature was screened and N = 247 articles were checked according to the PICOS criteria: MRI studies investigating transdiagnostic (across MDD, ANX, COM compared to healthy controls) and/or disorder-specific (between MDD, ANX, COM) neural alterations. N = 35, thereof n = 13 structural MRI and diffusion-tensor imaging studies and n = 22 functional MRI studies investigating emotional, cognitive deficits and resting state were included and quality coded.

RESULTS

Results indicated transdiagnostic structural/functional alterations in the orbitofrontal cortex/middle frontal cortex and in limbic regions (amygdala, cingulum, hippocampus). Few and inconsistent disorder-specific alterations were reported. However, depression-specific functional alterations were reported for the inferior frontal gyrus and dorsolateral prefrontal cortex during emotional tasks, and limbic regions at rest. Preliminary results for anxiety-specific functional alterations were found in the insula and frontal regions during emotional tasks, in the inferior parietal lobule, superior frontal gyrus and superior temporal gyrus during cognitive tasks, and (para)limbic alterations at rest.

CONCLUSIONS

This review provides evidence to support existing transdiagnostic fronto-limbic neural models in MDD and ANX. On top, it expands existing knowledge taking into account comorbidity and comparing MDD with ANX. Heterogeneous evidence exists for disorder-specific alterations. Research focusing on ANX sub-types, and the consideration of COM would contribute to a better understanding of basic neural underpinnings.

A review of systems biology research of anxiety disorders

The development of "omic" technologies and deep phenotyping may facilitate a systems biology approach to understanding anxiety disorders. Systems biology approaches incorporate data from multiple modalities (e.g., genomic, neuroimaging) with functional analyses (e.g., animal and tissue culture models) and mathematical modeling (e.g., machine learning) to investigate pathological biophysical networks at various scales. Here we review: i) the neurobiology of anxiety disorders; ii) how systems biology approaches have advanced this work; and iii) the clinical implications and future directions of this research. Systems biology approaches have provided an improved functional understanding of candidate biomarkers and have suggested future potential for refining the diagnosis, prognosis, and treatment of anxiety disorders. The systems biology approach for anxiety disorders is, however, in its infancy and in some instances is characterized by insufficient power and replication. The studies reviewed here represent important steps to further untangling the pathophysiology of anxiety disorders.

Panic attack prediction using wearable devices and machine learning: Development and cohort study (Preprint)

Background

A panic attack (PA) is an intense form of anxiety accompanied by multiple somatic presentations, leading to frequent emergency department visits and impairing the quality of life. A prediction model for PAs could help clinicians and patients monitor, control, and carry out early intervention for recurrent PAs, enabling more personalized treatment for panic disorder (PD).

Objective

This study aims to provide a 7-day PA prediction model and determine the relationship between a future PA and various features, including physiological factors, anxiety and depressive factors, and the air quality index (AQI).

Methods

We enrolled 59 participants with PD (Diagnostic and Statistical Manual of Mental Disorders, 5th edition, and the Mini International Neuropsychiatric Interview). Participants used smartwatches (Garmin Vívosmart 4) and mobile apps to collect their sleep, heart rate (HR), activity level, anxiety, and depression scores (Beck Depression Inventory [BDI], Beck Anxiety Inventory [BAI], State-Trait Anxiety Inventory state anxiety [STAI-S], State-Trait Anxiety Inventory trait anxiety [STAI-T], and Panic Disorder Severity Scale Self-Report) in their real life for a duration of 1 year. We also included AQIs from open data. To analyze these data, our team used 6 machine learning methods: random forests, decision trees, linear discriminant analysis, adaptive boosting, extreme gradient boosting, and regularized greedy forests.

Results

For 7-day PA predictions, the random forest produced the best prediction rate. Overall, the accuracy of the test set was 67.4%-81.3% for different machine learning algorithms. The most critical variables in the model were questionnaire and physiological features, such as the BAI, BDI, STAI, MINI, average HR, resting HR, and deep sleep duration.

Conclusions

It is possible to predict PAs using a combination of data from questionnaires and physiological and environmental data.

Biomarkers in Panic Disorder

Background:

Panic disorder (PD) is a kind of anxiety disorder that impacts the life quality and functional perspectives in patients. However, the pathophysiological study of PD seems still inadequate and many unresolved issues need to be clarified.

Objectives:

In this review article of biomarkers in PD, the investigator will focus on the findings of magnetic resonance imaging (MRI) of the brain in the pathophysiology study. The MRI biomarkers would be divided into several categories, on the basis of structural and functional perspectives.

Methods:

The structural category would include the gray matter and white matter tract studies. The functional category would consist of functional MRI (fMRI), resting-state fMRI (Rs-fMRI), and magnetic resonance spectroscopy (MRS). The PD biomarkers revealed by the above methodologies would be discussed in this article.

Results:

For the gray matter perspectives, the PD patients would have alterations in the volumes of fear network structures, such as the amygdala, parahippocampal gyrus, thalamus, anterior cingulate cortex, insula, and frontal regions. For the white matter tract studies, the PD patients seemed to have alterations in the fasciculus linking the fear network regions, such as the anterior thalamic radiation, uncinate fasciculus, fronto-occipital fasciculus, and superior longitudinal fasciculus. For the fMRI studies in PD, the significant results also focused on the fear network regions, such as the amygdala, hippocampus, thalamus, insula, and frontal regions. For the Rs-fMRI studies, PD patients seemed to have alterations in the regions of the default mode network and fear network model. At last, the MRS results showed alterations in neuron metabolites of the hippocampus, amygdala, occipital cortex, and frontal regions.

Conclusion:

The MRI biomarkers in PD might be compatible with the extended fear network model hypothesis in PD, which included the amygdala, hippocampus, thalamus, insula, frontal regions, and sensory-related cortex.

Identification of Major Psychiatric Disorders From Resting-State Electroencephalography Using a Machine Learning Approach

We aimed to develop a machine learning (ML) classifier to detect and compare major psychiatric disorders using electroencephalography (EEG). We retrospectively collected data from medical records, intelligence quotient (IQ) scores from psychological assessments, and quantitative EEG (QEEG) at resting-state assessments from 945 subjects [850 patients with major psychiatric disorders (six large-categorical and nine specific disorders) and 95 healthy controls (HCs)]. A combination of QEEG parameters including power spectrum density (PSD) and functional connectivity (FC) at frequency bands was used to establish models for the binary classification between patients with each disorder and HCs. The support vector machine, random forest, and elastic net ML methods were applied, and prediction performances were compared. The elastic net model with IQ adjustment showed the highest accuracy. The best feature combinations and classification accuracies for discrimination between patients and HCs with adjusted IQ were as follows: schizophrenia = alpha PSD, 93.83%; trauma and stress-related disorders = beta FC, 91.21%; anxiety disorders = whole band PSD, 91.03%; mood disorders = theta FC, 89.26%; addictive disorders = theta PSD, 85.66%; and obsessive-compulsive disorder = gamma FC, 74.52%. Our findings suggest that ML in EEG may predict major psychiatric disorders and provide an objective index of psychiatric disorders.

The Application of a Machine Learning-Based Brain Magnetic Resonance Imaging Approach in Major Depression

Major depressive disorder (MDD) shows a high prevalence and is associated with increased disability. While traditional studies aimed to investigate global characteristic neurobiological substrates of MDD, machine learning-based approaches focus on individual people rather than a group. Therefore, machine learning has been increasingly conducted and applied to clinical practice. Several previous neuroimaging studies used machine learning for stratifying MDD patients from healthy controls as well as in differentially diagnosing MDD apart from other psychiatric disorders. Also, machine learning has been used to predict treatment response using magnetic resonance imaging (MRI) results. Despite the recent accomplishments of machine learning-based MRI studies, small sample sizes and the heterogeneity of the depression group limit the generalizability of a machine learning-based predictive model. Future neuroimaging studies should integrate various materials such as genetic, peripheral, and clinical phenotypes for more accurate predictability of diagnosis and treatment response.

Predicting Response to Group Cognitive Behavioral Therapy in Asthma by a Small Number of Abnormal Resting-State Functional Connections

Group cognitive behavioral therapy (GCBT) is a successful psychotherapy for asthma. However, response varies considerably among individuals, and identifying biomarkers of GCBT has been challenging. Thus, the aim of this study was to predict an individual's potential response by using machine learning algorithms and functional connectivity (FC) and to improve the personalized treatment of GCBT. We use the lasso method to make the feature selection in the functional connections between brain regions, and we utilize t-test method to test the significant difference of these selected features. The feature selections are performed between controls (size = 20) and pre-GCBT patients (size = 20), pre-GCBT patients (size = 10) and post-GCBT patients (size = 10), and post-GCBT patients (size = 10) and controls (size = 10). Depending on these features, support vector classification was used to classify controls and pre- and post-GCBT patients. Pearson correlation analysis was employed to analyze the associations between clinical symptoms and the selected discriminated FCs in post-GCBT patients. At last, linear support vector regression was applied to predict the therapeutic effect of GCBT. After feature selection and significant analysis, five discriminated FC regarding neuroimaging biomarkers of GCBT were discovered, which are also correlated with clinical symptoms. Using these discriminated functional connections, we could accurately classify the patients before and after GCBT (classification accuracy, 80%) and predict the therapeutic effect of GCBT in asthma (predicted accuracy, 67.8%). The findings in this study would provide a novel sight toward GCBT response prediction and further confirm neural underpinnings of asthma. Moreover, our findings had clinical implications for personalized treatment by identifying asthmatic patients who will be appropriate for GCBT.

CLINICAL TRIAL REGISTRATION

The brain mechanisms of group cognitive behavioral therapy to improve the symptoms of asthma (Registration number: Chi-CTR-15007442, http://www.chictr.org.cn/index.aspx).

The Co-Morbidity between Bipolar and Panic Disorder in Fibromyalgia Syndrome

About half of the patients with fibromyalgia (FM) had a lifetime major depression episode and one third had a panic disorder (PD). Because the co-morbidity between bipolar disorder (BD) and PD marks a specific subtype of BD we aimed to investigate if co-morbid BD/PD (comBD/PD) occurs more frequently than the single disorder in FM patients and evaluate the clinical significance and timing of this co-morbidity. Further, we explored the role of co-morbid subthreshold BD and PD. In 118 patients with FM, lifetime threshold and sub-threshold mood disorders and PD were diagnosed with Diagnostic and Statistical Manual of Mental Disorders-IV-Text Revision (DSM-IV-TR) Clinical Interview. Demographic and clinical variables were compared in co-morbid BD/PD (comBD/PD) and not co-morbid BD/PD (nocomBD/PD) subgroups. The co-morbidity BD/PD was seen in 46.6% of FM patients and in 68.6% when patients with minor bipolar (MinBD) and sub-threshold panic were included. These rates are higher than those of the general population and BD outpatients. There were no statistically significant differences between threshold and sub-threshold comBD/PD and nocom-BD/PD subgroups in demographic and clinical parameters. In the majority of patients (78.2%), the onset of comBD/PD preceded or was contemporary with FM. These findings support the hypothesis that comBD/PD is related to the development of FM in a subgroup of patients.

Implementing Precision Psychiatry: A Systematic Review of Individualized Prediction Models for Clinical Practice

BACKGROUND

The impact of precision psychiatry for clinical practice has not been systematically appraised. This study aims to provide a comprehensive review of validated prediction models to estimate the individual risk of being affected with a condition (diagnostic), developing outcomes (prognostic), or responding to treatments (predictive) in mental disorders.

METHODS

PRISMA/RIGHT/CHARMS-compliant systematic review of the Web of Science, Cochrane Central Register of Reviews, and Ovid/PsycINFO databases from inception until July 21, 2019 (PROSPERO CRD42019155713) to identify diagnostic/prognostic/predictive prediction studies that reported individualized estimates in psychiatry and that were internally or externally validated or implemented. Random effect meta-regression analyses addressed the impact of several factors on the accuracy of prediction models.

FINDINGS

Literature search identified 584 prediction modeling studies, of which 89 were included. 10.4% of the total studies included prediction models internally validated (n = 61), 4.6% models externally validated (n = 27), and 0.2% (n = 1) models considered for implementation. Across validated prediction modeling studies (n = 88), 18.2% were diagnostic, 68.2% prognostic, and 13.6% predictive. The most frequently investigated condition was psychosis (36.4%), and the most frequently employed predictors clinical (69.5%). Unimodal compared to multimodal models (β = .29, P = .03) and diagnostic compared to prognostic (β = .84, p < .0001) and predictive (β = .87, P = .002) models were associated with increased accuracy.

INTERPRETATION

To date, several validated prediction models are available to support the diagnosis and prognosis of psychiatric conditions, in particular, psychosis, or to predict treatment response. Advancements of knowledge are limited by the lack of implementation research in real-world clinical practice. A new generation of implementation research is required to address this translational gap.

Neurobiology of the major psychoses: a translational perspective on brain structure and function-the FOR2107 consortium

Genetic (G) and environmental (E) factors are involved in the etiology and course of the major psychoses (MP), i.e. major depressive disorder (MDD), bipolar disorder (BD), schizoaffective disorder (SZA) and schizophrenia (SZ). The neurobiological correlates by which these predispositions exert their influence on brain structure, function and course of illness are poorly understood. In the FOR2107 consortium, animal models and humans are investigated. A human cohort of MP patients, healthy subjects at genetic and/or environmental risk, and control subjects (N = 2500) has been established. Participants are followed up after 2 years and twice underwent extensive deep phenotyping (MR imaging, clinical course, neuropsychology, personality, risk/protective factors, biomaterials: blood, stool, urine, hair, saliva). Methods for data reduction, quality assurance for longitudinal MRI data, and (deep) machine learning techniques are employed. In the parallelised animal cluster, genetic risk was introduced by a rodent model (Cacna1c deficiency) and its interactions with environmental risk and protective factors are studied. The animals are deeply phenotyped regarding cognition, emotion, and social function, paralleling the variables assessed in humans. A set of innovative experimental projects connect and integrate data from the human and animal parts, investigating the role of microRNA, neuroplasticity, immune signatures, (epi-)genetics and gene expression. Biomaterial from humans and animals are analyzed in parallel. The FOR2107 consortium will delineate pathophysiological entities with common neurobiological underpinnings ("biotypes") and pave the way for an etiologic understanding of the MP, potentially leading to their prevention, the prediction of individual disease courses, and novel therapies in the future.

Magnetic Resonance Imaging of Human Olfactory Dysfunction

Olfactory dysfunctions affect a larger portion of population (up to 15% with partial olfactory loss, and 5% with complete olfactory loss) as compared to other sensory dysfunctions (e.g. auditory or visual) and have a negative impact on the life quality. The impairment of olfactory functions may happen at each stage of the olfactory system, from epithelium to cortex. Non-invasive neuroimaging techniques such as the magnetic resonance imaging (MRI) have advanced the understanding of the advent and progress of olfactory dysfunctions in humans. The current review summarizes recent MRI studies on human olfactory dysfunction to present an updated and comprehensive picture of the structural and functional alterations in the central olfactory system as a consequence of olfactory loss and regain. Furthermore, the review also highlights recent progress on optimizing the olfactory functional MRI as well as new approaches for data processing that are promising for future clinical practice.

Machine learning in mental health: a scoping review of methods and applications

BACKGROUND

This paper aims to synthesise the literature on machine learning (ML) and big data applications for mental health, highlighting current research and applications in practice.

METHODS

We employed a scoping review methodology to rapidly map the field of ML in mental health. Eight health and information technology research databases were searched for papers covering this domain. Articles were assessed by two reviewers, and data were extracted on the article's mental health application, ML technique, data type, and study results. Articles were then synthesised via narrative review.

RESULTS

Three hundred papers focusing on the application of ML to mental health were identified. Four main application domains emerged in the literature, including: (i) detection and diagnosis; (ii) prognosis, treatment and support; (iii) public health, and; (iv) research and clinical administration. The most common mental health conditions addressed included depression, schizophrenia, and Alzheimer's disease. ML techniques used included support vector machines, decision trees, neural networks, latent Dirichlet allocation, and clustering.

CONCLUSIONS

Overall, the application of ML to mental health has demonstrated a range of benefits across the areas of diagnosis, treatment and support, research, and clinical administration. With the majority of studies identified focusing on the detection and diagnosis of mental health conditions, it is evident that there is significant room for the application of ML to other areas of psychology and mental health. The challenges of using ML techniques are discussed, as well as opportunities to improve and advance the field.

Abnormal frontal generator during auditory sensory gating in panic disorder: An MEG study

Patients with panic disorder (PD) exhibit abnormalities in early-stage information processing, even for the nonthreatening stimuli. A previous event-related potential study reported that PD patients show a deficit in sensory gating (SG), a protective mechanism of the brain to filter out irrelevant sensory inputs. However, there is no clear understanding about the neural correlates of SG deficits in PD. Moreover, whether SG deficits, if any, are associated with clinical manifestations remain unknown. In this study, 18 patients with PD and 20 age- and gender-matched healthy controls were recruited to perform auditory paired-stimulus paradigm using magnetoencephalographic (MEG) recordings. Results showed that PD patients demonstrated significantly higher M50 SG ratios in the right inferior frontal gyrus (RIFG) and higher M100 SG ratios in both RIFG and right superior temporal gyrus (RSTG) than those of the control group. It was important to note that in the RIFG, the M50 SG ratios correlated significantly with the scores of Body Sensation Questionnaire (BSQ) and Distractibility scale of Sensory Gating Inventory among patients with PD. In conclusion, this study suggests that PD patients exhibited a deficient ability to filter out irrelevant information, and such a defect might lead to cognitive misinterpretation of somatic sensations and distractibility.

Translational machine learning for psychiatric neuroimaging

Despite its initial promise, neuroimaging has not been widely translated into clinical psychiatry to assist in the prediction of diagnoses, prognoses, and optimal therapeutic strategies. Machine learning approaches may enhance the translational potential of neuroimaging because they specifically focus on overcoming biases by optimizing the generalizability of pipelines that measure complex brain patterns to predict targets at a single-subject level. This article introduces some fundamentals of a translational machine learning approach before selectively reviewing literature to-date. Promising initial results are then balanced by the description of limitations that should be considered in order to interpret existing research and maximize the possibility of future translation. Future directions are then presented in order to inspire further research and progress the field towards clinical translation.

The impact of depressive comorbidity on neural plasticity following cognitive-behavioral therapy in panic disorder with agoraphobia

BACKGROUND

Depressive disorders are a frequent comorbidity of panic disorder with agoraphobia (PD/AG). Cognitive-behavioral therapy (CBT) for PD/AG effectively reduces anxiety and depressive symptoms, irrespective of comorbidities. However, as depressive comorbidities can confound fear circuitry activation (i.e. amygdalae, insulae, anterior cingulate cortex) in PD/AG, we investigated whether comorbid depressive disorders alter neural plasticity following CBT.

METHODS

Within a randomized, controlled clinical trial on exposure-based CBT, forty-two PD/AG patients including fifteen (35.7%) with a comorbid depressive disorder (PD/AG + DEP) participated in a longitudinal functional magnetic resonance imaging (fMRI) study. A differential fear conditioning task was used as probe of interest. A generalized psycho-physiological interaction analysis (gPPI) served to study functional connectivity patterns.

RESULTS

After CBT, only PD/AG patients without comorbid depressive disorders (PD/AG-DEP) showed reduced activation in the left inferior frontal gyrus (IFG) extending to the insula. While PD/AG-DEP patients showed enhanced functional connectivity (FC) between the left IFG and subcortical structures (anterior cingulate cortex, thalamus and midbrain), PD/AG + DEP patients exhibited increased FC between the left IFG and cortical structures (prefrontal, parietal regions). In both groups, FC decreased following CBT.

LIMITATIONS

Primary depressed and medicated patients were excluded. Major depression and dysthymia were collapsed.

CONCLUSIONS

Reduced activation in the left IFG, as previously shown in PD/AG, appears to be a specific substrate of CBT effects in PD/AG-DEP patients only. Differential patterns of FC pertaining to fear circuitry networks in patients without depression vs. cognitive networks in patients with comorbid depression may point towards different pathways recruited by CBT as a function of comorbidity.

Digital Interventions for Mental Disorders: Key Features, Efficacy, and Potential for Artificial Intelligence Applications

Mental disorders are highly prevalent and often remain untreated. Many limitations of conventional face-to-face psychological interventions could potentially be overcome through Internet-based and mobile-based interventions (IMIs). This chapter introduces core features of IMIs, describes areas of application, presents evidence on the efficacy of IMIs as well as potential effect mechanisms, and delineates how Artificial Intelligence combined with IMIs may improve current practices in the prevention and treatment of mental disorders in adults. Meta-analyses of randomized controlled trials clearly show that therapist-guided IMIs can be highly effective for a broad range of mental health problems. Whether the effects of unguided IMIs are also clinically relevant, particularly under routine care conditions, is less clear. First studies on IMIs for the prevention of mental disorders have shown promising results. Despite limitations and challenges, IMIs are increasingly implemented into routine care worldwide. IMIs are also well suited for applications of Artificial Intelligence and Machine Learning, which provides ample opportunities to improve the identification and treatment of mental disorders. Together with methodological innovations, these approaches may also deepen our understanding of how psychological interventions work, and why. Ethical and professional restraints as well as potential contraindications of IMIs, however, should also be considered. In sum, IMIs have a high potential for improving the prevention and treatment of mental health disorders across various indications, settings, and populations. Therefore, implementing IMIs into routine care as both adjunct and alternative to face-to-face treatment is highly desirable. Technological advancements may further enhance the variability and flexibility of IMIs, and thus even further increase their impact in people's lives in the future.

Machine Learning Approaches for Clinical Psychology and Psychiatry

Machine learning approaches for clinical psychology and psychiatry explicitly focus on learning statistical functions from multidimensional data sets to make generalizable predictions about individuals. The goal of this review is to provide an accessible understanding of why this approach is important for future practice given its potential to augment decisions associated with the diagnosis, prognosis, and treatment of people suffering from mental illness using clinical and biological data. To this end, the limitations of current statistical paradigms in mental health research are critiqued, and an introduction is provided to critical machine learning methods used in clinical studies. A selective literature review is then presented aiming to reinforce the usefulness of machine learning methods and provide evidence of their potential. In the context of promising initial results, the current limitations of machine learning approaches are addressed, and considerations for future clinical translation are outlined.

A Conscious Resting State fMRI Study in SLE Patients Without Major Neuropsychiatric Manifestations

Neuropsychiatric systemic lupus erythematosus (NPSLE) is one of the main causes of death in patients with systemic lupus erythematosus (SLE). Signs and symptoms of NPSLE are heterogeneous, and it is hard to diagnose, and treat NPSLE patients in the early stage. We conducted this study to explore the possible brain activity changes using resting state functional magnetic resonance imaging (rs-fMRI) in SLE patients without major neuropsychiatric manifestations (non-NPSLE patients). We also tried to investigate the possible associations among brain activity, disease activity, depression, and anxiety. In our study, 118 non-NPSLE patients and 81 healthy controls (HC) were recruited. Rs-fMRI data were used to calculate the regional homogeneity (ReHo) in all participants. We found decreased ReHo values in the fusiform gyrus and thalamus and increased ReHo values in the parahippocampal gyrus and uncus. The disease activity was positively correlated with ReHo values of the cerebellum and negatively correlated with values in the frontal gyrus. Several brain areas showed correlations with depressive and anxiety statuses. These results suggested that abnormal brain activities might occur before NPSLE and might be the foundation of anxiety and depression symptoms. Early detection and proper treatment of brain dysfunction might prevent the progression to NPSLE. More studies are needed to understand the complicated underlying mechanisms.

Depression in patients with SAPHO syndrome and its relationship with brain activity and connectivity

BACKGROUND

Synovitis-acne-pustulosis-hyperostosis-osteitis (SAPHO) syndrome is a rare disease and there is no related literature concerning psychiatric symptoms in SAPHO patients. Thus, we believe that this will be the first paper to explore the episode and the neurobiological basis of depression symptoms in SAPHO patients using resting state functional magnetic resonance imaging (rs-fMRI). Twenty-eight SAPHO patients and fifteen age- and gender- matched normal controls (NC) were consecutively submitted to psychiatric evaluation and rs-fMRI scanning.

RESULTS

46.2% (13/28) of SAPHO patients were diagnosed as depression. The local spontaneous activity study showed that depressed SAPHO (D-SAPHO) patients had decreased amplitude of low-frequency fluctuation (ALFF) in the bilateral ventrolateral prefrontal cortex (VLPFC, attributed to the anatomical structures of Brodmann's area 47, 45 and 44) and right dorsolateral prefrontal cortex (DLPFC, attributed to the anatomical structures of Brodmann's area 8, 9 and 46), increased ALFF in the bilateral middle temporal gyrus, when compared to non-depressed SAPHO (ND-SAPHO) patients. The functional connectivity (FC) study disclosed that D-SAPHO patients had an increased FC in the anterior portions of default mode network (DMN) (the bilateral inferior frontal cortex, anterior cingulate cortex and insula cortex), and a decreased FC in the posterior areas of DMN (left middle occipital cortex), when compared to ND-SAPHO patients. Furthermore, correlation analysis revealed that both ALFF and FC values were significantly correlated with depression scores of SAPHO patients.

CONCLUSION

These results prompt us to understand the underlying pathophysiological mechanism of depression in SAPHO syndrome, and demonstrate that abnormal brain functional areas may serve as effective biological indicators to monitor depression in the future.

Support Vector Machine Analysis of Functional Magnetic Resonance Imaging of Interoception Does Not Reliably Predict Individual Outcomes of Cognitive Behavioral Therapy in Panic Disorder with Agoraphobia

BACKGROUND

The approach to apply multivariate pattern analyses based on neuro imaging data for outcome prediction holds out the prospect to improve therapeutic decisions in mental disorders. Patients suffering from panic disorder with agoraphobia (PD/AG) often exhibit an increased perception of bodily sensations. The purpose of this investigation was to assess whether multivariate classification applied to a functional magnetic resonance imaging (fMRI) interoception paradigm can predict individual responses to cognitive behavioral therapy (CBT) in PD/AG.

METHODS

This analysis is based on pretreatment fMRI data during an interoceptive challenge from a multicenter trial of the German PANIC-NET. Patients with DSM-IV PD/AG were dichotomized as responders (n = 30) or non-responders (n = 29) based on the primary outcome (Hamilton Anxiety Scale Reduction ≥50%) after 6 weeks of CBT (2 h/week). fMRI parametric maps were used as features for response classification with linear support vector machines (SVM) with or without automated feature selection. Predictive accuracies were assessed using cross validation and permutation testing. The influence of methodological parameters and the predictive ability for specific interoception-related symptom reduction were further evaluated.

RESULTS

SVM did not reach sufficient overall predictive accuracies (38.0-54.2%) for anxiety reduction in the primary outcome. In the exploratory analyses, better accuracies (66.7%) were achieved for predicting interoception-specific symptom relief as an alternative outcome domain. Subtle information regarding this alternative response criterion but not the primary outcome was revealed by post hoc univariate comparisons.

CONCLUSION

In contrast to reports on other neurofunctional probes, SVM based on an interoception paradigm was not able to reliably predict individual response to CBT. Results speak against the clinical applicability of this technique.

Functional neuroimaging of psychotherapeutic processes in anxiety and depression: from mechanisms to predictions

PURPOSE OF REVIEW

The review provides an update of functional neuroimaging studies that identify neural processes underlying psychotherapy and predict outcomes following psychotherapeutic treatment in anxiety and depressive disorders. Following current developments in this field, studies were classified as 'mechanistic' or 'predictor' studies (i.e., informing neurobiological models about putative mechanisms versus aiming to provide predictive information).

RECENT FINDINGS

Mechanistic evidence points toward a dual-process model of psychotherapy in anxiety disorders with abnormally increased limbic activation being decreased, while prefrontal activity is increased. Partly overlapping findings are reported for depression, albeit with a stronger focus on prefrontal activation following treatment. No studies directly comparing neural pathways of psychotherapy between anxiety and depression were detected. Consensus is accumulating for an overarching role of the anterior cingulate cortex in modulating treatment response across disorders. When aiming to quantify clinical utility, the need for single-subject predictions is increasingly recognized and predictions based on machine learning approaches show high translational potential.

SUMMARY

Present findings encourage the search for predictors providing clinically meaningful information for single patients. However, independent validation as a crucial prerequisite for clinical use is still needed. Identifying nonresponders a priori creates the need for alternative treatment options that can be developed based on an improved understanding of those neural mechanisms underlying effective interventions.