Separating generalized anxiety disorder from major depression using clinical, hormonal, and structural MRI data: A multimodal machine learning study

A novel approach to anxiety level prediction using small sets of judgment and survey variables

Anxiety, a condition characterized by intense fear and persistent worry, affects millions each year and, when severe, is distressing and functionally impairing. Numerous machine learning frameworks have been developed and tested to predict features of anxiety and anxiety traits. This study extended these approaches by using a small set of interpretable judgment variables (n = 15) and contextual variables (demographics, perceived loneliness, COVID-19 history) to (1) understand the relationships between these variables and (2) develop a framework to predict anxiety levels [derived from the State Trait Anxiety Inventory (STAI)]. This set of 15 judgment variables, including loss aversion and risk aversion, models biases in reward/aversion judgments extracted from an unsupervised, short (2-3 min) picture rating task (using the International Affective Picture System) that can be completed on a smartphone. The study cohort consisted of 3476 de-identified adult participants from across the United States who were recruited using an email survey database. Using a balanced Random Forest approach with these judgment and contextual variables, STAI-derived anxiety levels were predicted with up to 81% accuracy and 0.71 AUC ROC. Normalized Gini scores showed that the most important predictors (age, loneliness, household income, employment status) contributed a total of 29-31% of the cumulative relative importance and up to 61% was contributed by judgment variables. Mediation/moderation statistics revealed that the interactions between judgment and contextual variables appears to be important for accurately predicting anxiety levels. Median shifts in judgment variables described a behavioral profile for individuals with higher anxiety levels that was characterized by less resilience, more avoidance, and more indifference behavior. This study supports the hypothesis that distinct constellations of 15 interpretable judgment variables, along with contextual variables, could yield an efficient and highly scalable system for mental health assessment. These results contribute to our understanding of underlying psychological processes that are necessary to characterize what causes variance in anxiety conditions and its behaviors, which can impact treatment development and efficacy.

Early identification of Parkinson's disease with anxiety based on combined clinical and MRI features

Objective

To identify cortical and subcortical volume, thickness and cortical area features and the networks they constituted related to anxiety in Parkinson's disease (PD) using structural magnetic resonance imaging (sMRI), and to integrate multimodal features based on machine learning to identify PD-related anxiety.

Methods

A total of 219 patients with PD were retrospectively enrolled in the study. 291 sMRI features including cortical volume, subcortical volume, cortical thickness, and cortical area, as well as 17 clinical features, were extracted. Graph theory analysis was used to explore structural networks. A support vector machine (SVM) combination model, which used both sMRI and clinical features to identify participants with PD-related anxiety, was developed and evaluated. The performance of SVM models were evaluated. The mean impact value (MIV) of the feature importance evaluation algorithm was used to rank the relative importance of sMRI features and clinical features within the model.

Results

17 significant sMRI variables associated with PD-related anxiety was used to build a brain structural network. And seven sMRI and 5 clinical features with statistically significant differences were incorporated into the SVM model. The comprehensive model achieved higher performance than clinical features or sMRI features did alone, with an accuracy of 0.88, a precision of 0.86, a sensitivity of 0.81, an F1-Score of 0.83, a macro-average of 0.85, a weighted-average of 0.92, an AUC of 0.88, and a result of 10-fold cross-validation of 0.91 in test set. The sMRI feature right medialorbitofrontal thickness had the highest impact on the prediction model.

Conclusion

We identified the brain structural features and networks related to anxiety in PD, and developed and internally validated a comprehensive model with multimodal features in identifying.

Mood Disorder Severity and Subtype Classification Using Multimodal Deep Neural Network Models

The subtype diagnosis and severity classification of mood disorder have been made through the judgment of verified assistance tools and psychiatrists. Recently, however, many studies have been conducted using biomarker data collected from subjects to assist in diagnosis, and most studies use heart rate variability (HRV) data collected to understand the balance of the autonomic nervous system on statistical analysis methods to perform classification through statistical analysis. In this research, three mood disorder severity or subtype classification algorithms are presented through multimodal analysis of data on the collected heart-related data variables and hidden features from the variables of time and frequency domain of HRV. Comparing the classification performance of the statistical analysis widely used in existing major depressive disorder (MDD), anxiety disorder (AD), and bipolar disorder (BD) classification studies and the multimodality deep neural network analysis newly proposed in this study, it was confirmed that the severity or subtype classification accuracy performance of each disease improved by 0.118, 0.231, and 0.125 on average. Through the study, it was confirmed that deep learning analysis of biomarker data such as HRV can be applied as a primary identification and diagnosis aid for mental diseases, and that it can help to objectively diagnose psychiatrists in that it can confirm not only the diagnosed disease but also the current mood status.

Classification of suicidality by training supervised machine learning models with brain MRI findings: A systematic review

BACKGROUND

Suicide is a global public health issue causing around 700,000 deaths worldwide each year. Therefore, identifying suicidal thoughts and behaviors in patients can help lower the suicide-related mortality rate. This review aimed to investigate the feasibility of suicidality identification by applying supervised Machine Learning (ML) methods to Magnetic Resonance Imaging (MRI) data.

METHODS

We conducted a systematic search on PubMed, Scopus, and Web of Science to identify studies examining suicidality by applying ML methods to MRI features. Also, the Prediction Model Risk of Bias Assessment Tool (PROBAST) was employed for the quality assessment.

RESULTS

23 studies met the inclusion criteria. Of these, 20 developed prediction models without external validation and 3 developed prediction models with external validation. The performance of ML models varied among the reviewed studies, with the highest reported values of accuracies and Area Under the Curve (AUC) ranging from 51.7 % to 100 % and 0.52 to 1, respectively. Over half of the studies that reported accuracy (12/21) or AUC (13/16) achieved values of ≥0.8. Our comparative analysis indicated that deep learning exhibited the highest predictive performance compared to other ML models. The most commonly identified discriminative imaging features were resting-state functional connectivity and grey matter volume within prefrontal-limbic structures.

LIMITATIONS

Small sample sizes, lack of external validation, heterogeneous study designs, and ML model development.

CONCLUSIONS

Most of the studies developed ML models capable of ML-based suicide identification, although ML models' predictive performance varied across the reviewed studies. Thus, further well-designed is necessary to uncover the true potential of different ML models in this field.

Neuroimaging Study of Brain Functional Differences in Generalized Anxiety Disorder and Depressive Disorder

Generalized anxiety disorder (GAD) and depressive disorder (DD) are distinct mental disorders, which are characterized by complex and unique neuroelectrophysiological mechanisms in psychiatric neurosciences. The understanding of the brain functional differences between GAD and DD is crucial for the accurate diagnosis and clinical efficacy evaluation. The aim of this study was to reveal the differences in functional brain imaging between GAD and DD based on multidimensional electroencephalogram (EEG) characteristics. To this end, 10 min resting-state EEG signals were recorded from 38 GAD and 34 DD individuals. Multidimensional EEG features were subsequently extracted, which include power spectrum density (PSD), fuzzy entropy (FE), and phase lag index (PLI). Then, a direct statistical analysis (i.e., ANOVA) and three ensemble learning models (i.e., Random Forest (RF), Light Gradient Boosting Machine (LightGBM), eXtreme Gradient Boosting (XGBoost)) were used on these EEG features for the differential recognitions. Our results showed that DD has significantly higher PSD values in the alpha1 and beta band, and a higher FE in the beta band, in comparison with GAD, along with the aberrant functional connections in all four bands between GAD and DD. Moreover, machine learning analysis further revealed that the distinct features predominantly occurred in the beta band and functional connections. Here, we show that DD has higher power and more complex brain activity patterns in the beta band and reorganized brain functional network structures in all bands compared to GAD. In sum, these findings move towards the practical identification of brain functional differences between GAD and DD.

Feasibility of applying graph theory to diagnosing generalized anxiety disorder using machine learning models

The aim of this study was to investigate whether the alterations of topological properties can facilitate the diagnosis of generalized anxiety disorder (GAD). Twenty first-episode drug-naive Chinese individuals with GAD and twenty age-sex-education-matched healthy controls (HCs) were included in the primary training set, and the results of which were validated using nineteen drug-free patients with GAD and nineteen unmatched HCs. Two 3 T scanners were used to acquire T1, diffusion tensor, and resting-state functional images. Topological properties were altered in the functional cerebral networks among patients with GAD, but not in the structural networks. Using the nodal topological properties in the anti-correlated functional networks, machine learning models distinguished drug-naive GADs from their matched HCs independent of the type of kernels and the amount of features. Although the models built with drug-naive GADs failed to distinguish drug-free GADs from HCs, the features selected for those models could be used to build new models for distinguishing drug-free GADs from HCs. Our findings suggested that it is feasible to utilize the topological characteristics of brain network to facilitate the diagnosis of GAD. However, further research with decent sample sizes, multimodal features, and improved modeling methods are needed to build more robust models.

Alterations in regional homogeneity and functional connectivity associated with cognitive impairment in patients with hypertension: a resting-state functional magnetic resonance imaging study

Our study aims to investigate the alterations and diagnostic efficiency of regional homogeneity (ReHo) and functional connectivity (FC) in hypertension patients with cognitive impairment. A total of 62 hypertension patients with cognitive impairment (HTN-CI), 59 hypertension patients with normal cognition (HTN-NC), and 58 healthy controls (HCs) with rs-fMRI data were enrolled in this study. Univariate analysis (based on whole-brain ReHo and seed-based FC maps) was performed to observe brain regions with significant differences among the three groups. Multiple voxel pattern analysis (MVPA) was applied to evaluate the diagnostic accuracy in classifying HTN-CI from HTN-NC and HCs. Compared with the HCs and HTN-NC, HTN-CI exhibited decreased ReHo in the right caudate, left postcentral gyrus, posterior cingulate gyrus, insula, while increased ReHo in the left superior occipital gyrus and superior parietal gyrus. HTN-CI showed increased FC between seed regions (left posterior cingulate gyrus, insula, postcentral gyrus) with many specific brain regions. MVPA analysis (based on whole-brain ReHo and seed-based FC maps) displayed high classification ability in distinguishing HTN-CI from HTN-NC and HCs. The ReHo values (right caudate) and the FC values (left postcentral gyrus seed to left posterior cingulate gyrus) were positively correlated with the MoCA scores in HTN-CI. HTN-CI was associated with decreased ReHo and increased FC mainly in the left posterior cingulate gyrus, postcentral gyrus, insula compared to HTN-NC and HC. Besides, MVPA analysis yields excellent diagnostic accuracy in classifying HTN-CI from HTN-NC and HCs. The findings may contribute to unveiling the underlying neuropathological mechanism of HTN-CI.

A Review of Machine Learning and Deep Learning Approaches on Mental Health Diagnosis

Combating mental illnesses such as depression and anxiety has become a global concern. As a result of the necessity for finding effective ways to battle these problems, machine learning approaches have been included in healthcare systems for the diagnosis and probable prediction of the treatment outcomes of mental health conditions. With the growing interest in machine and deep learning methods, analysis of existing work to guide future research directions is necessary. In this study, 33 articles on the diagnosis of schizophrenia, depression, anxiety, bipolar disorder, post-traumatic stress disorder (PTSD), anorexia nervosa, and attention deficit hyperactivity disorder (ADHD) were retrieved from various search databases using the preferred reporting items for systematic reviews and meta-analysis (PRISMA) review methodology. These publications were chosen based on their use of machine learning and deep learning technologies, individually assessed, and their recommended methodologies were then classified into the various disorders included in this study. In addition, the difficulties encountered by the researchers are discussed, and a list of some public datasets is provided.

A Novel Approach for Developing Efficient and Convenient Short Assessments to Approximate a Long Assessment

This paper describes a novel Long to Short approach that uses machine learning to develop efficient and convenient short assessments to approximate a long assessment. This approach is applicable to any assessments used to assess people's behaviors, opinions, attitudes, mental and physical states, traits, aptitudes, abilities, and mastery of a subject matter. We demonstrated the Long to Short approach on the Depression Anxiety Stress Scale (DASS-42) for assessing anxiety levels in adults. We first obtained data for the original assessment from a large sample of participants. We then derived the total scores from participants' responses to all items of the long assessment as the ground truths. Next, we used feature selection techniques to select participants' responses to a subset of items of the long assessment to predict the ground truths accurately. We then trained machine learning models that uses the minimal number of items needed to achieve the prediction accuracy similar to that when the responses to all items of the whole long assessment are used. We generated all possible combinations of minimal number of items to create multiple short assessments of similar predictive accuracies for use if the short assessment is to be done repeatedly. Finally, we implemented the short anxiety assessments in a web application for convenient use with any future participant of the assessment.

Multimodal machine learning in precision health: A scoping review

Machine learning is frequently being leveraged to tackle problems in the health sector including utilization for clinical decision-support. Its use has historically been focused on single modal data. Attempts to improve prediction and mimic the multimodal nature of clinical expert decision-making has been met in the biomedical field of machine learning by fusing disparate data. This review was conducted to summarize the current studies in this field and identify topics ripe for future research. We conducted this review in accordance with the PRISMA extension for Scoping Reviews to characterize multi-modal data fusion in health. Search strings were established and used in databases: PubMed, Google Scholar, and IEEEXplore from 2011 to 2021. A final set of 128 articles were included in the analysis. The most common health areas utilizing multi-modal methods were neurology and oncology. Early fusion was the most common data merging strategy. Notably, there was an improvement in predictive performance when using data fusion. Lacking from the papers were clear clinical deployment strategies, FDA-approval, and analysis of how using multimodal approaches from diverse sub-populations may improve biases and healthcare disparities. These findings provide a summary on multimodal data fusion as applied to health diagnosis/prognosis problems. Few papers compared the outputs of a multimodal approach with a unimodal prediction. However, those that did achieved an average increase of 6.4% in predictive accuracy. Multi-modal machine learning, while more robust in its estimations over unimodal methods, has drawbacks in its scalability and the time-consuming nature of information concatenation.

Gray and white matter structural examination for diagnosis of major depressive disorder and subthreshold depression in adolescents and young adults: a preliminary radiomics analysis

BACKGROUND

Radiomics is an emerging image analysis framework that provides more details than conventional methods. In present study, we aimed to identify structural radiomics features of gray matter (GM) and white matter (WM), and to develop and validate the classification model for major depressive disorder (MDD) and subthreshold depression (StD) diagnosis using radiomics analysis.

METHODS

A consecutive cohort of 142 adolescents and young adults, including 43 cases with MDD, 49 cases with StD and 50 healthy controls (HC), were recruited and underwent the three-dimensional T1 weighted imaging (3D-T1WI) and diffusion tensor imaging (DTI). We extracted radiomics features representing the shape and diffusion properties of GM and WM from all participants. Then, an all-relevant feature selection process embedded in a 10-fold cross-validation framework was used to identify features with significant power for discrimination. Random forest classifiers (RFC) were established and evaluated successively using identified features.

RESULTS

The results showed that a total of 3030 features were extracted after preprocessing, including 2262 shape-related features from each T1-weighted image representing GM morphometry and 768 features from each DTI representing the diffusion properties of WM. 25 features were selected ultimately, including ten features for MDD versus HC, eight features for StD versus HC, and seven features for MDD versus StD. The accuracies and area under curve (AUC) the RFC achieved were 86.75%, 0.93 for distinguishing MDD from HC with significant radiomics features located in the left medial orbitofrontal cortex, right superior and middle temporal regions, right anterior cingulate, left cuneus and hippocampus, 70.51%, 0.69 for discriminating StD from HC within left cuneus, medial orbitofrontal cortex, cerebellar vermis, hippocampus, anterior cingulate and amygdala, right superior and middle temporal regions, and 59.15%, 0.66 for differentiating MDD from StD within left medial orbitofrontal cortex, middle temporal and cuneus, right superior frontal, superior temporal regions and hippocampus, anterior cingulate, respectively.

CONCLUSION

These findings provide preliminary evidence that radiomics features of brain structure are valid for discriminating MDD and StD subjects from healthy controls. The MRI-based radiomics approach, with further improvement and validation, might be a potential facilitating method to clinical diagnosis of MDD or StD.

Predictive utility of symptom measures in classifying anxiety and depression: A machine-learning approach

Major depressive disorder (MDD) and generalized anxiety disorder (GAD) are highly prevalent, co-occurring disorders with significant symptom overlap, posing challenges in accurately distinguishing and diagnosing these disorders. The tripartite model proposes that anxious arousal is specific to anxiety and anhedonia is specific to depression, though anxious apprehension may play a greater role in GAD than anxious arousal. The present study tested the efficacy of the Mood and Anxiety Symptom Questionnaire anhedonic depression (MASQ-AD) and anxious arousal (MASQ-AA) scales and the Penn State Worry Questionnaire (PSWQ) in identifying lifetime or current MDD, current major depressive episode (MDE), and GAD using binary support vector machine learning algorithms in an adult sample (n = 150). The PSWQ and MASQ-AD demonstrated predictive utility in screening for and identification of GAD and current MDE respectively, with the MASQ-AD eight-item subscale outperforming the MASQ-AD 14-item subscale. The MASQ-AA did not predict MDD, current MDE, or GAD, and the MASQ-AD did not predict current or lifetime MDD. The PSWQ and MASQ-AD are efficient and accurate screening tools for GAD and current MDE. Results support the tripartite model in that anhedonia is unique to depression, but inclusion of anxious apprehension as a separate dimension of anxiety is warranted.

An insight into diagnosis of depression using machine learning techniques: a systematic review

BACKGROUND

In this modern era, depression is one of the most prevalent mental disorders from which millions of individuals are affected today. The symptoms of depression are heterogeneous and often coincide with other disorders such as bipolar disorder, Parkinson's, schizophrenia, etc. It is a serious mental illness that may lead to other health problems if left untreated. Currently, identifying individuals with depression is totally based on the expertise of the clinician's experience. In order to assist clinicians in identifying the characteristics and classifying depressed people, different types of data modalities and machine learning techniques have been incorporated by researchers in this field. This study aims to find the answers to some important questions related to the trend of publications, data modality, machine learning models, dataset usage, pre-processing techniques and feature extraction and selection techniques that are prevalent and guide the direction of future research on depression diagnosis.

METHODS

This systematic review was conducted using a broad range of articles from two major databases: IEEE Xplore and PubMed. Studies ranging from the years 2011 to April 2021 were retrieved from the databases resulting in a total of 590 articles (53 articles from the IEEE Xplore database and 537 articles from the PubMed database). Out of those, the articles which satisfied the defined inclusion criteria were investigated for further analysis.

RESULTS

A total of 135 articles were identified and analysed for this review. High growth in the number of publications has been observed in recent years. Furthermore, significant diversity in the use of data modalities and machine learning classifiers has also been noted in this study. fMRI data with an SVM classifier was found to be the most popular choice among researchers. In most of the studies, data scarcity and small sample size, particularly for neuroimaging data are major concerns. The use of identical data pre-processing tools for similar data modalities can be seen. This study also provides statistical analysis of the current framework with respect to the modality, machine learning classifier, sample size and accuracy by applying one-way ANOVA and the Tukey - Kramer test.

CONCLUSION

The results indicate that an effective fusion of machine learning techniques with a potential data modality has a promising future for assisting clinicians in automatic depression diagnosis.

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.

A neural network approach to optimising treatments for depression using data from specialist and community psychiatric services in Australia, New Zealand and Japan

This study investigated the application of a recurrent neural network for optimising pharmacological treatment for depression. A clinical dataset of 458 participants from specialist and community psychiatric services in Australia, New Zealand and Japan were extracted from an existing custom-built, web-based tool called Psynary . This data, which included baseline and self-completed reviews, was used to train and refine a novel algorithm which was a fully connected network feature extractor and long short-term memory algorithm was firstly trained in isolation and then integrated and annealed using slow learning rates due to the low dimensionality of the data. The accuracy of predicting depression remission before processing patient review data was 49.8%. After processing only 2 reviews, the accuracy was 76.5%. When considering a change in medication, the precision of changing medications was 97.4% and the recall was 71.4% . The medications with predicted best results were antipsychotics (88%) and selective serotonin reuptake inhibitors (87.9%). This is the first study that has created an all-in-one algorithm for optimising treatments for all subtypes of depression. Reducing treatment optimisation time for patients suffering with depression may lead to earlier remission and hence reduce the high levels of disability associated with the condition. Furthermore, in a setting where mental health conditions are increasing strain on mental health services, the utilisation of web-based tools for remote monitoring and machine/deep learning algorithms may assist clinicians in both specialist and primary care in extending specialist mental healthcare to a larger patient community.

Mental Health Prediction Using Machine Learning: Taxonomy, Applications, and Challenges

The increase of mental health problems and the need for effective medical health care have led to an investigation of machine learning that can be applied in mental health problems. This paper presents a recent systematic review of machine learning approaches in predicting mental health problems. Furthermore, we will discuss the challenges, limitations, and future directions for the application of machine learning in the mental health field. We collect research articles and studies that are related to the machine learning approaches in predicting mental health problems by searching reliable databases. Moreover, we adhere to the PRISMA methodology in conducting this systematic review. We include a total of 30 research articles in this review after the screening and identification processes. Then, we categorize the collected research articles based on the mental health problems such as schizophrenia, bipolar disorder, anxiety and depression, posttraumatic stress disorder, and mental health problems among children. Discussing the findings, we reflect on the challenges and limitations faced by the researchers on machine learning in mental health problems. Additionally, we provide concrete recommendations on the potential future research and development of applying machine learning in the mental health field.

Predicting the naturalistic course in anxiety disorders using clinical and biological markers: a machine learning approach

BACKGROUND

Disease trajectories of patients with anxiety disorders are highly diverse and approximately 60% remain chronically ill. The ability to predict disease course in individual patients would enable personalized management of these patients. This study aimed to predict recovery from anxiety disorders within 2 years applying a machine learning approach.

METHODS

In total, 887 patients with anxiety disorders (panic disorder, generalized anxiety disorder, agoraphobia, or social phobia) were selected from a naturalistic cohort study. A wide array of baseline predictors (N = 569) from five domains (clinical, psychological, sociodemographic, biological, lifestyle) were used to predict recovery from anxiety disorders and recovery from all common mental disorders (CMDs: anxiety disorders, major depressive disorder, dysthymia, or alcohol dependency) at 2-year follow-up using random forest classifiers (RFCs).

RESULTS

At follow-up, 484 patients (54.6%) had recovered from anxiety disorders. RFCs achieved a cross-validated area-under-the-receiving-operator-characteristic-curve (AUC) of 0.67 when using the combination of all predictor domains (sensitivity: 62.0%, specificity 62.8%) for predicting recovery from anxiety disorders. Classification of recovery from CMDs yielded an AUC of 0.70 (sensitivity: 64.6%, specificity: 62.3%) when using all domains. In both cases, the clinical domain alone provided comparable performances. Feature analysis showed that prediction of recovery from anxiety disorders was primarily driven by anxiety features, whereas recovery from CMDs was primarily driven by depression features.

CONCLUSIONS

The current study showed moderate performance in predicting recovery from anxiety disorders over a 2-year follow-up for individual patients and indicates that anxiety features are most indicative for anxiety improvement and depression features for improvement in general.

Machine learning approaches for parsing comorbidity/heterogeneity in antisociality and substance use disorders: A primer

By some accounts, as many as 93% of individuals diagnosed with antisocial personality disorder (ASPD) or psychopathy also meet criteria for some form of substance use disorder (SUD). This high level of comorbidity, combined with an overlapping biopsychosocial profile, and potentially interacting features, has made it difficult to delineate the shared/unique characteristics of each disorder. Moreover, while rarely acknowledged, both SUD and antisociality exist as highly heterogeneous disorders in need of more targeted parcellation. While emerging data-driven nosology for psychiatric disorders (e.g., Research Domain Criteria (RDoC), Hierarchical Taxonomy of Psychopathology (HiTOP)) offers the opportunity for a more systematic delineation of the externalizing spectrum, the interrogation of large, complex neuroimaging-based datasets may require data-driven approaches that are not yet widely employed in psychiatric neuroscience. With this in mind, the proposed article sets out to provide an introduction into machine learning methods for neuroimaging that can help parse comorbid, heterogeneous externalizing samples. The modest machine learning work conducted to date within the externalizing domain demonstrates the potential utility of the approach but remains highly nascent. Within the paper, we make suggestions for how future work can make use of machine learning methods, in combination with emerging psychiatric nosology systems, to further diagnostic and etiological understandings of the externalizing spectrum. Finally, we briefly consider some challenges that will need to be overcome to encourage further progress in the field.

Prediction of anxiety disorders using a feature ensemble based bayesian neural network

Anxiety disorders are common among youth, posing risks to physical and mental health development. Early screening can help identify such disorders and pave the way for preventative treatment. To this end, the Youth Online Diagnostic Assessment (YODA) tool was developed and deployed to predict youth disorders using online screening questionnaires filled by parents. YODA facilitated collection of several novel unique datasets of self-reported anxiety disorder symptoms. Since the data is self-reported and often noisy, feature selection needs to be performed on the raw data to improve accuracy. However, a single set of selected features may not be informative enough. Consequently, in this work we propose and evaluate a novel feature ensemble based Bayesian Neural Network (FE-BNN) that exploits an ensemble of features for improving the accuracy of disorder predictions. We evaluate the performance of FE-BNN on three disorder-specific datasets collected by YODA. Our method achieved the AUC of 0.8683, 0.8769, 0.9091 for the predictions of Separation Anxiety Disorder, Generalized Anxiety Disorder and Social Anxiety Disorder, respectively. These results provide initial evidence that our method outperforms the original diagnostic scoring function of YODA and several other baseline methods for three anxiety disorders, which can practically help prioritizing diagnostic interviews. Our promising results call for investigation of interpretable methods maintaining high predictive accuracy.

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.

Application of Artificial Intelligence in the MRI Classification Task of Human Brain Neurological and Psychiatric Diseases: A Scoping Review

Artificial intelligence (AI) for medical imaging is a technology with great potential. An in-depth understanding of the principles and applications of magnetic resonance imaging (MRI), machine learning (ML), and deep learning (DL) is fundamental for developing AI-based algorithms that can meet the requirements of clinical diagnosis and have excellent quality and efficiency. Moreover, a more comprehensive understanding of applications and opportunities would help to implement AI-based methods in an ethical and sustainable manner. This review first summarizes recent research advances in ML and DL techniques for classifying human brain magnetic resonance images. Then, the application of ML and DL methods to six typical neurological and psychiatric diseases is summarized, including Alzheimer's disease (AD), Parkinson's disease (PD), major depressive disorder (MDD), schizophrenia (SCZ), attention-deficit/hyperactivity disorder (ADHD), and autism spectrum disorder (ASD). Finally, the limitations of the existing research are discussed, and possible future research directions are proposed.

Structural brain correlates in major depression, anxiety disorders and post-traumatic stress disorder: A voxel-based morphometry meta-analysis

The high comorbidity of Major Depressive Disorder (MDD), Anxiety Disorders (ANX), and Posttraumatic Stress Disorder (PTSD) has hindered the study of their structural neural correlates. The authors analyzed specific and common grey matter volume (GMV) characteristics by comparing them with healthy controls (HC). The meta-analysis of voxel-based morphometry (VBM) studies showed unique GMV diminutions for each disorder (p < 0.05, corrected) and less robust smaller GMV across diagnostics (p < 0.01, uncorrected). Pairwise comparison between the disorders showed GMV differences in MDD versus ANX and in ANX versus PTSD. These results endorse the hypothesis that unique clinical features characterizing MDD, ANX, and PTSD are also reflected by disorder specific GMV correlates.

Fear-induced brain activations distinguish anxious and trauma-exposed brains

Translational models of fear conditioning and extinction have elucidated a core neural network involved in the learning, consolidation, and expression of conditioned fear and its extinction. Anxious or trauma-exposed brains are characterized by dysregulated neural activations within regions of this fear network. In this study, we examined how the functional MRI activations of 10 brain regions commonly activated during fear conditioning and extinction might distinguish anxious or trauma-exposed brains from controls. To achieve this, activations during four phases of a fear conditioning and extinction paradigm in 304 participants with or without a psychiatric diagnosis were studied. By training convolutional neural networks (CNNs) using task-specific brain activations, we reliably distinguished the anxious and trauma-exposed brains from controls. The performance of models decreased significantly when we trained our CNN using activations from task-irrelevant brain regions or from a brain network that is irrelevant to fear. Our results suggest that neuroimaging data analytics of task-induced brain activations within the fear network might provide novel prospects for development of brain-based psychiatric diagnosis.

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.

Machine Learning Prediction of Treatment Outcome in Late-Life Depression

Background: Recent evidence suggests that integration of multi-modal data improves performance in machine learning prediction of depression treatment outcomes. Here, we compared the predictive performance of three machine learning classifiers using differing combinations of sociodemographic characteristics, baseline clinical self-reports, cognitive tests, and structural magnetic resonance imaging (MRI) features to predict treatment outcomes in late-life depression (LLD). Methods: Data were combined from two clinical trials conducted with depressed adults aged 60 and older, including response to escitalopram (N = 32, NCT01902004) and Tai Chi (N = 35, NCT02460666). Remission was defined as a score of 6 or less on the 24-item Hamilton Rating Scale for Depression (HAMD) at the end of 24 weeks of treatment. Features subsets were constructed from baseline sociodemographic and clinical features, gray matter volumes (GMVs), or both. Three classification algorithms were compared: (1) Support Vector Machine-Radial Bias Function (SVMRBF), (2) Random Forest (RF), and (3) Logistic Regression (LR). A repeated 5-fold cross-validation approach with a wrapper-based feature selection method was used for model fitting. Model performance metrics included Area under the ROC Curve (AUC) and Matthews correlation coefficient (MCC). Cross-validated performance significance was tested by permutation analysis. Classifiers were compared by Cochran's Q and post-hoc pairwise comparisons using McNemar's Chi-Square test with Bonferroni correction. Results: For the RF and SVMRBF algorithms, the combined feature set outperformed the clinical and GMV feature sets with a final cross-validated AUC of 0.83 ± 0.11 and 0.80 ± 0.11, respectively. Both classifiers passed permutation analysis. The LR algorithm performed best using GMV features alone (AUC 0.79 ± 0.14) but failed to pass permutation analysis using any feature set. Performance of the three classifiers differed significantly for all three features sets. Important predictive features of treatment response included anterior and posterior cingulate volumes, depression characteristics, and self-reported health-related quality scores. Conclusion: This preliminary exploration into the use of ML and multi-modal data to identify predictors of general treatment response in LLD indicates that integration of clinical and structural MRI features significantly increases predictive capability. Identified features are among those previously implicated in geriatric depression, encouraging future work in this arena.

Towards a brain-based predictome of mental illness

Neuroimaging-based approaches have been extensively applied to study mental illness in recent years and have deepened our understanding of both cognitively healthy and disordered brain structure and function. Recent advancements in machine learning techniques have shown promising outcomes for individualized prediction and characterization of patients with psychiatric disorders. Studies have utilized features from a variety of neuroimaging modalities, including structural, functional, and diffusion magnetic resonance imaging data, as well as jointly estimated features from multiple modalities, to assess patients with heterogeneous mental disorders, such as schizophrenia and autism. We use the term "predictome" to describe the use of multivariate brain network features from one or more neuroimaging modalities to predict mental illness. In the predictome, multiple brain network-based features (either from the same modality or multiple modalities) are incorporated into a predictive model to jointly estimate features that are unique to a disorder and predict subjects accordingly. To date, more than 650 studies have been published on subject-level prediction focusing on psychiatric disorders. We have surveyed about 250 studies including schizophrenia, major depression, bipolar disorder, autism spectrum disorder, attention-deficit hyperactivity disorder, obsessive-compulsive disorder, social anxiety disorder, posttraumatic stress disorder, and substance dependence. In this review, we present a comprehensive review of recent neuroimaging-based predictomic approaches, current trends, and common shortcomings and share our vision for future directions.

Toward Robust Anxiety Biomarkers: A Machine Learning Approach in a Large-Scale Sample

BACKGROUND

The field of psychiatry has long sought biomarkers that can objectively diagnose patients, predict treatment response, or identify individuals at risk of illness onset. However, reliable psychiatric biomarkers have yet to emerge. The recent application of machine learning techniques to develop neuroimaging-based biomarkers has yielded promising preliminary results. However, much of the work in this domain has not met best practice standards from the field of machine learning. This is especially true for studies of anxiety, creating uncertainty about the potential for anxiety biomarker development.

METHODS

We applied machine learning tools to predict trait anxiety from neuroimaging measurements in humans. Using publicly available data from the Brain Genomics Superstruct Project, we compared a suite of neuroimaging-based machine learning models predicting anxiety within a discovery sample (n = 531, 307 women) via k-fold cross-validation, and we tested the final model (a stacked model incorporating region-to-region functional connectivity, amygdala seed-to-voxel connectivity, and volumetric and cortical thickness data) in a held-out, unseen test sample (n = 348, 209 women).

RESULTS

Though the best model was able to predict anxiety within the discovery sample (cross-validated R2 of .06, permutation test p < .001), the generalization test within the holdout sample failed (R2 of -.04, permutation test p > .05).

CONCLUSIONS

In this study, we did not find evidence of a generalizable anxiety biomarker. However, we encourage other researchers to investigate this topic, utilizing large samples and proper methodology, to clarify the potential of neuroimaging-based anxiety biomarkers.

Towards Precision Medicine in Psychosis: Benefits and Challenges of Multimodal Multicenter Studies-PSYSCAN: Translating Neuroimaging Findings From Research into Clinical Practice

In the last 2 decades, several neuroimaging studies investigated brain abnormalities associated with the early stages of psychosis in the hope that these could aid the prediction of onset and clinical outcome. Despite advancements in the field, neuroimaging has yet to deliver. This is in part explained by the use of univariate analytical techniques, small samples and lack of statistical power, lack of external validation of potential biomarkers, and lack of integration of nonimaging measures (eg, genetic, clinical, cognitive data). PSYSCAN is an international, longitudinal, multicenter study on the early stages of psychosis which uses machine learning techniques to analyze imaging, clinical, cognitive, and biological data with the aim of facilitating the prediction of psychosis onset and outcome. In this article, we provide an overview of the PSYSCAN protocol and we discuss benefits and methodological challenges of large multicenter studies that employ neuroimaging measures.

Common and different neural markers in major depression and anxiety disorders: A pilot structural magnetic resonance imaging study

Although anxiety and depression often co-occur and share some clinical features, it is still unclear if they are neurobiologically distinct or similar processes. In this study, we explored common and specific cortical morphology alterations in depression and anxiety disorders. Magnetic Resonance Imaging data were acquired from 13 Major Depressive Disorder (MDD), 11 Generalized Anxiety Disorder (GAD), 11 Panic Disorder (PD) patients and 21 healthy controls (HC). Regional cortical thickness, surface area (SA), volume and gyrification were measured and compared among groups. We found left orbitofrontal thinning in all patient groups, as well as disease-specific alterations. MDD showed volume deficits in left precentral gyrus compared to all groups, volume and area deficits in right fusiform gyrus compared to GAD and HC. GAD showed lower SA than MDD and PD in right superior parietal cortex, higher gyrification than HC in right frontal gyrus. PD showed higher gyrification in left superior parietal cortex when compared to MDD and higher SA in left postcentral gyrus compared to all groups. Our results suggest that clinical phenotypic similarities between major depression and anxiety disorders might rely on common prefrontal alterations. Frontotemporal and parietal abnormalities may represent unique biological signatures of depression and anxiety.

Machine learning and big data: Implications for disease modeling and therapeutic discovery in psychiatry

INTRODUCTION

Machine learning capability holds promise to inform disease models, the discovery and development of novel disease modifying therapeutics and prevention strategies in psychiatry. Herein, we provide an introduction on how machine learning/Artificial Intelligence (AI) may instantiate such capabilities, as well as provide rationale for its application to psychiatry in both research and clinical ecosystems.

METHODS

Databases PubMed and PsycINFO were searched from 1966 to June 2016 for keywords:Big Data, Machine Learning, Precision Medicine, Artificial Intelligence, Mental Health, Mental Disease, Psychiatry, Data Mining, RDoC, and Research Domain Criteria. Articles selected for review were those that were determined to be aligned with the objective of this particular paper.

RESULTS

Results indicate that AI is a viable option to build useful predictors of outcome while offering objective and comparable accuracy metrics, a unique opportunity, particularly in mental health research. The approach has also consistently brought notable insight into disease models through processing the vast amount of already available multi-domain, semi-structured medical data. The opportunity for AI in psychiatry, in addition to disease-model refinement, is in characterizing those at risk, and it is likely also relevant to personalizing and discovering therapeutics.

CONCLUSIONS

Machine learning currently provides an opportunity to parse disease models in complex, multi-factorial disease states (e.g. mental disorders) and could possibly inform treatment selection with existing therapies and provide bases for domain-based therapeutic discovery.

Data Analytics in Smart Healthcare: The Recent Developments and Beyond

The concepts of the smart city and the Internet of Things (IoT) have been facilitating the rollout of medical devices and systems to capture valuable information of humanity. A lot of artificial intelligence techniques have been demonstrated to be effective in smart city applications like energy, transportation, retail and control. In recent decade, retardation of the adoption of data analytics algorithms and systems in healthcare has been decreasing, and there is tremendous growth in data analytics research on healthcare data. The results of analytics aim at improving people’s quality of life as well as relieving the issue of medical shortages. In this special issue “Data Analytics in Smart Healthcare”, thirteen (13) papers have been published as the representative examples of recent developments. Guest Editors also highlight some emergent topics and opening challenges in healthcare analytics which follow the visions of the movement of healthcare analytics research.

Secretory Acid Sphingomyelinase in the Serum of Medicated Patients Predicts the Prospective Course of Depression

Major depressive disorder (MDD) is a highly prevalent and devastating psychiatric illness with strong individual and societal burdens. However, biomarkers to improve the limited preventive and therapeutic approaches are scarce. Multilevel evidence suggests that the pathophysiological involvement of sphingolipids particularly increases the levels of ceramides and the ceramide hydrolyzing enzyme, acid sphingomyelinase. The activity of secretory acid sphingomyelinase (S-ASM) and routine blood parameters were determined in the serum of patients with current (unmedicated n = 63, medicated n = 66) and remitted (n = 39) MDD and healthy subjects (n = 61). Depression severity and anxiety and their 3-weeks prospective course of treatment were assessed by psychometric inventories. S-ASM activity was not different between the four groups, did not decrease during treatment, and was not lower in individuals taking medication that functionally inhibited ASM. However, S-ASM correlated positively with depression severity only in remitted patients. High enzyme activity at inclusion predicted milder clinician-evaluated and self-rated depression severity (HAM-D, MADRS, BDI-II) and state anxiety at follow-up, and was related to stronger improvement in these scores in medicated patients. S-ASM was strongly and contrariwise associated with serum lipids in unmedicated and medicated females. These findings contribute to a better understanding of the pathomechanisms underlying depression and the development of clinical strategies and biomarkers.

An automatic diagnostic network using skew-robust adversarial discriminative domain adaptation to evaluate the severity of depression

BACKGROUND AND OBJECTIVE

Deep learning provides an automatic and robust solution to depression severity evaluation. However, despite it is powerful, there is a trade-off between robust performance and the cost of manual annotation.

METHODS

Motivated by knowledge evolution and domain adaptation, we propose a deep evaluation network using skew-robust adversarial discriminative domain adaptation (SRADDA), which adaptively shifts its domain from a large-scale Twitter dataset to a small-scale depression interview dataset for evaluating the severity of depression.

RESULTS

Without top-down selection, SRADDA-based severity evaluation network achieves regression errors of 6.38 (Root Mean Square Error,RMSE) and 4.93 (Mean Absolute Error,MAE), which outperforms baselines provided by the Audio/Visual Emotion Challenge and Workshop(AVEC 2017). However, with top-down selection, the network achieves comparable results (RMSE = 5.13, MAE = 4.08).

CONCLUSIONS

Results show that SRADDA not only represents features robustly, but also performs comparably to state-of-the-art results on small-scale dataset, DAIC-WOZ.

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.

Development of Neuroimaging-Based Biomarkers in Psychiatry

This chapter presents an overview of accumulating neuroimaging data with emphasis on translational potential. The subject will be described in the context of three disease states, i.e., schizophrenia, bipolar disorder, and major depressive disorder, and for three clinical goals, i.e., disease risk assessment, subtyping, and treatment decision.

Machine learning in major depression: From classification to treatment outcome prediction

AIMS

Major depression disorder (MDD) is the single greatest cause of disability and morbidity, and affects about 10% of the population worldwide. Currently, there are no clinically useful diagnostic biomarkers that are able to confirm a diagnosis of MDD from bipolar disorder (BD) in the early depressive episode. Therefore, exploring translational biomarkers of mood disorders based on machine learning is in pressing need, though it is challenging, but with great potential to improve our understanding of these disorders.

DISCUSSIONS

In this study, we review popular machine-learning methods used for brain imaging classification and predictions, and provide an overview of studies, specifically for MDD, that have used magnetic resonance imaging data to either (a) classify MDDs from controls or other mood disorders or (b) investigate treatment outcome predictors for individual patients. Finally, challenges, future directions, and potential limitations related to MDD biomarker identification are also discussed, with a goal of offering a comprehensive overview that may help readers to better understand the applications of neuroimaging data mining in depression.

CONCLUSIONS

We hope such efforts may highlight the need for an urgently needed paradigm shift in treatment, to guide personalized optimal clinical care.

Frequency-Specific Functional Connectivity Density as an Effective Biomarker for Adolescent Generalized Anxiety Disorder

Several neuropsychiatric diseases have been found to influence the frequency-specific spontaneous functional brain organization (SFBO) in resting state, demonstrating that the abnormal brain activities of different frequency bands are associated with various physiological and psychological dysfunctions. However, little is known about the frequency specificities of SFBO in adolescent generalized anxiety disorder (GAD). Here, a novel complete ensemble empirical mode decomposition with adaptive noise method was applied to decompose the time series of each voxel across all participants (31 adolescent patients with GAD and 28 matched healthy controls; HCs) into four frequency-specific bands with distinct intrinsic oscillation. The functional connectivity density (FCD) of different scales (short-range and long-range) was calculated to quantify the SFBO changes related to GAD within each above frequency-specific band and the conventional frequency band (0.01–0.08 Hz). Support vector machine classifier was further used to examine the discriminative ability of the frequency-specific FCD values. The results showed that adolescent GAD patients exhibited abnormal alterations of both short-range and long-range FCD (S-FCD and L-FCD) in widespread brain regions across three frequency-specific bands. Positive correlation between the State Anxiety Inventory (SAI) score and increased L-FCD in the fusiform gyrus in the conventional frequency band was found in adolescents with GAD. Both S-FCD and L-FCD in the insula in the lower frequency band (0.02–0.036 Hz) had the highest classification performance compared to all other brain regions with inter-group difference. Furthermore, a satisfactory classification performance was achieved by combining the discrepant S-FCD and L-FCD values in all frequency bands, with the area under the curve (AUC) value of 0.9414 and the corresponding sensitivity, specificity, and accuracy of 87.15, 92.92, and 89.83%, respectively. This study indicates that the alterations of SFBO in adolescent GAD are frequency dependence and the frequency-specific FCD can potentially serve as a valuable biomarker in discriminating GAD patients from HCs. These findings may provide new insights into the pathophysiological mechanisms of adolescent GAD.

Developmental relations between amygdala volume and anxiety traits: Effects of informant, sex, and age

Although substantial human and animal evidence suggests a role for the amygdala in anxiety, literature linking amygdala volume to anxiety symptomatology is inconclusive, with studies finding positive, negative, and null results. Clarifying this brain–behavior relation in middle to late childhood is especially important, as this is a time both of amygdala structural maturation and the emergence of many anxiety disorders. The goal of the current study was to clarify inconsistent findings in previous literature by identifying factors moderating the relation between amygdala volume and anxiety traits in a large sample of typically developing children aged 6–13 years (N = 72). In particular, we investigated the moderating effects of informant (parent vs. child), age, and sex. We found that children's reports (i.e., self-reports) were related to amygdala volume; children who reported higher anxiety levels had smaller amygdalae. This negative relation between amygdala volume and anxiety weakened with age. There was also an independent effect of sex, such that relations were stronger in males than in females. These results indicate the importance of considering sample and informant characteristics when charting the neurobiological mechanisms underlying developmental anxiety.

Separating generalized anxiety disorder from major depression using clinical, hormonal, and structural MRI data: A multimodal machine learning study

BACKGROUND

Generalized anxiety disorder (GAD) is difficult to recognize and hard to separate from major depression (MD) in clinical settings. Biomarkers might support diagnostic decisions. This study used machine learning on multimodal biobehavioral data from a sample of GAD, MD and healthy subjects to differentiate subjects with a disorder from healthy subjects (case-classification) and to differentiate GAD from MD (disorder-classification).

METHODS

Subjects with GAD (n = 19), MD without GAD (n = 14), and healthy comparison subjects (n = 24) were included. The sample was matched regarding age, sex, handedness and education and free of psychopharmacological medication. Binary support vector machines were used within a nested leave-one-out cross-validation framework. Clinical questionnaires, cortisol release, gray matter (GM), and white matter (WM) volumes were used as input data separately and in combination.

RESULTS

Questionnaire data were well-suited for case-classification but not disorder-classification (accuracies: 96.40%, p < .001; 56.58%, p > .22). The opposite pattern was found for imaging data (case-classification GM/WM: 58.71%, p = .09/43.18%, p > .66; disorder-classification GM/WM: 68.05%, p = .034/58.27%, p > .15) and for cortisol data (38.02%, p = .84; 74.60%, p = .009). All data combined achieved 90.10% accuracy (p < .001) for case-classification and 67.46% accuracy (p = .0268) for disorder-classification.

CONCLUSIONS

In line with previous evidence, classification of GAD was difficult using clinical questionnaire data alone. Particularly cortisol and GM volume data were able to provide incremental value for the classification of GAD. Findings suggest that neurobiological biomarkers are a useful target for further research to delineate their potential contribution to diagnostic processes.