Quantitative prediction of individual psychopathology in trauma survivors using resting-state FMRI

Exploring the impact of biological sex on intrinsic connectivity networks in PTSD: A data-driven approach

INTRODUCTION

Sex as a biological variable (SABV) may help to account for the differential development and expression of post-traumatic stress disorder (PTSD) symptoms among trauma-exposed males and females. Here, we investigate the impact of SABV on PTSD-related neural alterations in resting-state functional connectivity (rsFC) within three core intrinsic connectivity networks (ICNs): the salience network (SN), central executive network (CEN), and default mode network (DMN).

METHODS

Using an independent component analysis (ICA), we compared rsFC of the SN, CEN, and DMN between males and females, with and without PTSD (n = 47 females with PTSD, n = 34 males with PTSD, n = 36 healthy control females, n = 20 healthy control males) via full factorial ANCOVAs. Additionally, linear regression analyses were conducted with clinical variables (i.e., PTSD and depression symptoms, childhood trauma scores) in order to determine intrinsic network connectivity characteristics specific to SABV. Furthermore, we utilized machine learning classification models to predict the biological sex and PTSD diagnosis of individual participants based on intrinsic network activity patterns.

RESULTS

Our findings revealed differential network connectivity patterns based on SABV and PTSD diagnosis. Males with PTSD exhibited increased intra-SN (i.e., SN-anterior insula) rsFC and increased DMN-right superior parietal lobule/precuneus/superior occipital gyrus rsFC as compared to females with PTSD. There were also differential network connectivity patterns for comparisons between the PTSD and healthy control groups for males and females, separately. We did not observe significant correlations between clinical measures of interest and brain region clusters which displayed significant between group differences as a function of biological sex, thus further reinforcing that SABV analyses are likely not confounded by these variables. Furthermore, machine learning classification models accurately predicted biological sex and PTSD diagnosis among novel/unseen participants based on ICN activation patterns.

CONCLUSION

This study reveals groundbreaking insights surrounding the impact of SABV on PTSD-related ICN alterations using data-driven methods. Our discoveries contribute to further defining neurobiological markers of PTSD among females and males and may offer guidance for differential sex-related treatment needs.

Unveiling the core functional networks of cognition: An ontology-guided machine learning approach

Deciphering the functional architecture that underpins diverse cognitive functions is fundamental quest in neuroscience. In this study, we employed an innovative machine learning framework that integrated cognitive ontology with functional connectivity analysis to identify brain networks essential for cognition. We identified a core assembly of functional connectomes, primarily located within the association cortex, which showed superior predictive performance compared to two conventional methods widely employed in previous research across various cognitive domains. Our approach achieved a mean prediction accuracy of 0.13 across 16 cognitive tasks, including working memory, reading comprehension, and sustained attention, outperforming the traditional methods' accuracy of 0.08. In contrast, our method showed limited predictive power for sensory, motor, and emotional functions, with a mean prediction accuracy of 0.03 across 9 relevant tasks, slightly lower than the traditional methods' accuracy of 0.04. These cognitive connectomes were further characterized by distinctive patterns of resting-state functional connectivity, structural connectivity via white matter tracts, and gene expression, highlighting their neurogenetic underpinnings. Our findings reveal a domain-general functional network fingerprint that pivotal to cognition, offering a novel computational approach to explore the neural foundations of cognitive abilities.

Unveiling the Core Functional Networks of Cognition: An Ontology-Guided Machine Learning Approach

Deciphering the functional architecture that underpins diverse cognitive functions is fundamental quest in neuroscience. In this study, we employed an innovative machine learning framework that integrated cognitive ontology with functional connectivity analysis to identify brain networks essential for cognition. We identified a core assembly of functional connectomes, primarily located within the association cortex, which showed superior predictive performance compared to two conventional methods widely employed in previous research across various cognitive domains. Our approach achieved a mean prediction accuracy of 0.13 across 16 cognitive tasks, including working memory, reading comprehension, and sustained attention, outperforming the traditional methods' accuracy of 0.08. In contrast, our method showed limited predictive power for sensory, motor, and emotional functions, with a mean prediction accuracy of 0.03 across 9 relevant tasks, slightly lower than the traditional methods' accuracy of 0.04. These cognitive connectomes were further characterized by distinctive patterns of resting-state functional connectivity, structural connectivity via white matter tracts, and gene expression, highlighting their neurogenetic underpinnings. Our findings reveal a domain-general functional network fingerprint that pivotal to cognition, offering a novel computational approach to explore the neural foundations of cognitive abilities.

Connectivity-based neuromarker for children's inhibitory control ability and its relevance to body mass index

Preserving a normal body mass index (BMI) is crucial for the healthy growth and development of children. As a core aspect of executive functions, inhibitory control plays a pivotal role in maintaining a normal BMI, which is key to preventing issues of childhood obesity. By studying individual variations in inhibitory control performance and its associated connectivity-based neuromarker in a sample of primary school students (N = 64; 9-12 yr), we aimed to unravel the pathway through which inhibitory control impacts children's BMI. Utilizing resting-state functional MRI scans and a connectivity-based psychometric prediction framework, we found that enhanced inhibitory control abilities were primarily associated with increased functional connectivity in brain structures vital to executive functions, such as the superior frontal lobule, superior parietal lobule, and posterior cingulate cortex. Conversely, inhibitory control abilities displayed a negative relationship with functional connectivity originating from reward-related brain structures, such as the orbital frontal and ventral medial prefrontal lobes. Furthermore, we revealed that both inhibitory control and its corresponding neuromarker can moderate the association between food-related delayed gratification and BMI in children. However, only the neuromarker of inhibitory control maintained its moderating effect on children's future BMI, as determined in the follow-up after one year. Overall, our findings shed light on the potential mechanisms of how inhibitory control in children impacts BMI, highlighting the utility of the connectivity-based neuromarker of inhibitory control in the context of childhood obesity.

Neural correlates of PTSD in women with childhood sexual abuse with and without PTSD and response to paroxetine treatment: A placebo-controlled, double-blind trial

Objective

Childhood sexual abuse is the leading cause of posttraumatic stress disorder (PTSD) in women, and is a prominent cause of morbidity and loss of function for which limited treatments are available. Understanding the neurobiology of treatment response is important for developing new treatments. The purpose of this study was to assess neural correlates of personalized traumatic memories in women with childhood sexual abuse with and without PTSD, and to assess response to treatment.

Methods

Women with childhood sexual abuse with (N = 28) and without (N = 17) PTSD underwent brain imaging with High-Resolution Positron Emission Tomography scanning with radiolabeled water for brain blood flow measurements during exposure to personalized traumatic scripts and memory encoding tasks. Women with PTSD were randomized to paroxetine or placebo followed by three months of double-blind treatment and repeat imaging with the same protocol.

Results

Women with PTSD showed decreases in areas involved in the Default Mode Network (DMN), a network of brain areas usually active when the brain is at rest, hippocampus and visual processing areas with exposure to traumatic scripts at baseline while women without PTSD showed increased activation in superior frontal gyrus and other areas (p < 0.005). Treatment of women with PTSD with paroxetine resulted in increased anterior cingulate activation and brain areas involved in the DMN and visual processing with scripts compared to placebo (p < 0.005).

Conclusion

PTSD related to childhood sexual abuse in women is associated with alterations in brain areas involved in memory and the stress response and treatment with paroxetine results in modulation of these areas.

Machine learning applied to fMRI patterns of brain activation in response to mutilation pictures predicts PTSD symptoms

BACKGROUND

The present study aimed to apply multivariate pattern recognition methods to predict posttraumatic stress symptoms from whole-brain activation patterns during two contexts where the aversiveness of unpleasant pictures was manipulated by the presence or absence of safety cues.

METHODS

Trauma-exposed participants were presented with neutral and mutilation pictures during functional magnetic resonance imaging (fMRI) collection. Before the presentation of pictures, a text informed the subjects that the pictures were fictitious ("safe context") or real-life scenes ("real context"). We trained machine learning regression models (Gaussian process regression (GPR)) to predict PTSD symptoms in real and safe contexts.

RESULTS

The GPR model could predict PTSD symptoms from brain responses to mutilation pictures in the real context but not in the safe context. The brain regions with the highest contribution to the model were the occipito-parietal regions, including the superior parietal gyrus, inferior parietal gyrus, and supramarginal gyrus. Additional analysis showed that GPR regression models accurately predicted clusters of PTSD symptoms, nominal intrusion, avoidance, and alterations in cognition. As expected, we obtained very similar results as those obtained in a model predicting PTSD total symptoms.

CONCLUSION

This study is the first to show that machine learning applied to fMRI data collected in an aversive context can predict not only PTSD total symptoms but also clusters of PTSD symptoms in a more aversive context. Furthermore, this approach was able to identify potential biomarkers for PTSD, especially in occipitoparietal regions.

Predicting sex, age, general cognition and mental health with machine learning on brain structural connectomes

There is an increasing expectation that advanced, computationally expensive machine learning (ML) techniques, when applied to large population-wide neuroimaging datasets, will help to uncover key differences in the human brain in health and disease. We take a comprehensive approach to explore how multiple aspects of brain structural connectivity can predict sex, age, general cognitive function and general psychopathology, testing different ML algorithms from deep learning (DL) model (BrainNetCNN) to classical ML methods. We modelled N = 8183 structural connectomes from UK Biobank using six different structural network weightings obtained from diffusion MRI. Streamline count generally provided the highest prediction accuracies in all prediction tasks. DL did not improve on prediction accuracies from simpler linear models. Further, high correlations between gradient attribution coefficients from DL and model coefficients from linear models suggested the models ranked the importance of features in similar ways, which indirectly suggested the similarity in models' strategies for making predictive decision to some extent. This highlights that model complexity is unlikely to improve detection of associations between structural connectomes and complex phenotypes with the current sample size.

Towards assessing subcortical "deep brain" biomarkers of PTSD with functional near-infrared spectroscopy

Assessment of brain function with functional near-infrared spectroscopy (fNIRS) is limited to the outer regions of the cortex. Previously, we demonstrated the feasibility of inferring activity in subcortical "deep brain" regions using cortical functional magnetic resonance imaging (fMRI) and fNIRS activity in healthy adults. Access to subcortical regions subserving emotion and arousal using affordable and portable fNIRS is likely to be transformative for clinical diagnostic and treatment planning. Here, we validate the feasibility of inferring activity in subcortical regions that are central to the pathophysiology of posttraumatic stress disorder (PTSD; i.e. amygdala and hippocampus) using cortical fMRI and simulated fNIRS activity in a sample of adolescents diagnosed with PTSD (N = 20, mean age = 15.3 ± 1.9 years) and age-matched healthy controls (N = 20, mean age = 14.5 ± 2.0 years) as they performed a facial expression task. We tested different prediction models, including linear regression, a multilayer perceptron neural network, and a k-nearest neighbors model. Inference of subcortical fMRI activity with cortical fMRI showed high prediction performance for the amygdala (r > 0.91) and hippocampus (r > 0.95) in both groups. Using fNIRS simulated data, relatively high prediction performance for deep brain regions was maintained in healthy controls (r > 0.79), as well as in youths with PTSD (r > 0.75). The linear regression and neural network models provided the best predictions.

Episodic Memory-Related Imaging Features as Valuable Biomarkers for the Diagnosis of Alzheimer's Disease: A Multicenter Study Based on Machine Learning

BACKGROUND

Individualized and reliable biomarkers are crucial for diagnosing Alzheimer's disease (AD). However, lack of accessibility and neurobiological correlation are the main obstacles to their clinical application. Machine learning algorithms can effectively identify personalized biomarkers based on the prominent symptoms of AD.

METHODS

Episodic memory-related magnetic resonance imaging (MRI) features of 143 patients with amnesic mild cognitive impairment (MCI) were identified using a multivariate relevance vector regression algorithm. The support vector machine classification model was constructed using these MRI features and verified in 2 independent datasets (N = 994). The neurobiological basis was also investigated based on cognitive assessments, neuropathologic biomarkers of cerebrospinal fluid, and positron emission tomography images of amyloid-β plaques.

RESULTS

The combination of gray matter volume and amplitude of low-frequency fluctuation MRI features accurately predicted episodic memory impairment in individual patients with amnesic MCI (r = 0.638) when measured using an episodic memory assessment panel. The MRI features that contributed to episodic memory prediction were primarily distributed across the default mode network and limbic network. The classification model based on these features distinguished patients with AD from normal control subjects with more than 86% accuracy. Furthermore, most identified episodic memory-related regions showed significantly different amyloid-β positron emission tomography measurements among the AD, MCI, and normal control groups. Moreover, the classification outputs significantly correlated with cognitive assessment scores and cerebrospinal fluid pathological biomarkers' levels in the MCI and AD groups.

CONCLUSIONS

Neuroimaging features can reflect individual episodic memory function and serve as potential diagnostic biomarkers of AD.

Prediction of individual trigeminal pain sensitivity from gray matter structure within the sensorimotor network

OBJECTIVE

To determine whether multivariate pattern regression analysis based on gray matter (GM) images constrained to the sensorimotor network could accurately predict trigeminal heat pain sensitivity in healthy individuals.

BACKGROUND

Prediction of individual pain sensitivity is of clinical relevance as high pain sensitivity is associated with increased risks of postoperative pain, pain chronification, and a poor treatment response. However, as pain is a subjective experience accurate identification of such individuals can be difficult. GM structure of sensorimotor regions have been shown to vary with pain sensitivity. It is unclear whether GM structure within these regions can be used to predict pain sensitivity.

METHODS

In this cross-sectional study, structural magnetic resonance images and pain thresholds in response to contact heat stimulation of the left supraorbital area were obtained from 79 healthy participants. Voxel-based morphometry was used to extract segmented and normalized GM images. These were then constrained to a mask encompassing the functionally defined resting-state sensorimotor network. The masked images and pain thresholds entered a multivariate relevance vector regression analysis for quantitative prediction of the individual pain thresholds. The correspondence between predicted and actual pain thresholds was indexed by the Pearson correlation coefficient (r) and the mean squared error (MSE). The generalizability of the model was assessed by 10-fold and 5-fold cross-validation. Non-parametric permutation tests were used to estimate significance levels.

RESULTS

Trigeminal heat pain sensitivity could be predicted from GM structure within the sensorimotor network with significant accuracy (10-fold: r = 0.53, p < 0.001, MSE = 10.32, p = 0.001; 5-fold: r = 0.46, p = 0.001, MSE = 10.54, p < 0.001). The resulting multivariate weight maps revealed that accurate prediction relied on multiple widespread regions within the sensorimotor network.

CONCLUSION

A multivariate pattern of GM structure within the sensorimotor network could be used to make accurate predictions about trigeminal heat pain sensitivity at the individual level in healthy participants. Widespread regions within the sensorimotor network contributed to the predictive model.

Prediction of Facial Emotion Recognition Ability in Patients With First-Episode Schizophrenia Using Amplitude of Low-Frequency Fluctuation-Based Support Vector Regression Model

Objective

There were few studies that had attempted to predict facial emotion recognition (FER) ability at the individual level in schizophrenia patients. In this study, we developed a model for the prediction of FER ability in Chinese Han patients with the first-episode schizophrenia (FSZ).

Materials and Methods

A total of 28 patients with FSZ and 33 healthy controls (HCs) were recruited. All subjects underwent resting-state fMRI (rs-fMRI). The amplitude of low-frequency fluctuation (ALFF) method was selected to analyze voxel-level spontaneous neuronal activity. The visual search experiments were selected to evaluate the FER, while the support vector regression (SVR) model was selected to develop a model based on individual rs-fMRI brain scan.

Results

Group difference in FER ability showed statistical significance (P < 0.05). In FSZ patients, increased mALFF value were observed in the limbic lobe and frontal lobe, while decreased mALFF value were observed in the frontal lobe, parietal lobe, and occipital lobe (P < 0.05, AlphaSim correction). SVR analysis showed that abnormal spontaneous activity in multiple brain regions, especially in the right posterior cingulate, right precuneus, and left calcarine could effectively predict fearful FER accuracy (r = 0.64, P = 0.011) in patients.

Conclusion

Our study provides an evidence that abnormal spontaneous activity in specific brain regions may serve as a predictive biomarker for fearful FER ability in schizophrenia.

Hippocampal Resting-State Functional Connectivity Forecasts Individual Posttraumatic Stress Disorder Symptoms: A Data-Driven Approach

BACKGROUND

Posttraumatic stress disorder (PTSD) is a debilitating disorder, and there is no current accurate prediction of who develops it after trauma. Neurobiologically, individuals with chronic PTSD exhibit aberrant resting-state functional connectivity (rsFC) between the hippocampus and other brain regions (e.g., amygdala, prefrontal cortex, posterior cingulate), and these aberrations correlate with severity of illness. Previous small-scale research (n < 25) has also shown that hippocampal rsFC measured acutely after trauma is predictive of future severity using a region-of-interest-based approach. While this is a promising biomarker, to date, no study has used a data-driven approach to test whole-brain hippocampal FC patterns in forecasting the development of PTSD symptoms.

METHODS

A total of 98 adults at risk of PTSD were recruited from the emergency department after traumatic injury and completed resting-state functional magnetic resonance imaging (8 min) within 1 month; 6 months later, they completed the Clinician-Administered PTSD Scale for DSM-5 for assessment of PTSD symptom severity. Whole-brain rsFC values with bilateral hippocampi were extracted (using CONN) and used in a machine learning kernel ridge regression analysis (PRoNTo); a k-folds (k = 10) and 70/30 testing versus training split approach were used for cross-validation (1000 iterations to bootstrap confidence intervals for significance values).

RESULTS

Acute hippocampal rsFC significantly predicted Clinician-Administered PTSD Scale for DSM-5 scores at 6 months (r = 0.30, p = .006; mean squared error = 120.58, p = .006; R² = 0.09, p = .025). In post hoc analyses, hippocampal rsFC remained significant after controlling for demographics, PTSD symptoms at baseline, and depression, anxiety, and stress severity at 6 months (B = 0.59, SE = 0.20, p = .003).

CONCLUSIONS

Findings suggest that functional connectivity of the hippocampus across the brain acutely after traumatic injury is associated with prospective PTSD symptom severity.

Functional near-infrared spectroscopy brain imaging predicts symptom severity in youth exposed to traumatic stress

Functional near-infrared spectroscopy (fNIRS) is a non-invasive neuroimaging technique with the potential to enable the assessment of posttraumatic stress disorder (PTSD) brain biomarkers in an affordable and portable manner. Consistent with biological models of PTSD, functional magnetic resonance imaging (fMRI) and fNIRS studies of adults with trauma exposure and PTSD symptoms suggest increased activation in the dorsolateral prefrontal cortex (dlPFC) and ventrolateral PFC (vlPFC) in response to negative emotion stimuli. We tested this theory with fNIRS assessment among youth exposed to traumatic stress and experiencing PTSD symptoms (PTSS). A portable fNIRS system collected hemodynamic responses from (N = 57) youth with PTSS when engaging in a classic emotion expression task that included fearful and neutral faces stimuli. The General Linear Model was applied to identify cortical activations associated with the facial stimuli. Subsequently, a prediction model was established via a Support Vector Regression to determine whether PTSS severity could be predicted based on fNIRS-derived cortical response measures and individual demographic information. Results were consistent with findings from adult fMRI and fNIRS studies of PTSS showing increased activation in the dlPFC and vlPFC in response to negative emotion stimuli. Subsequent prediction analysis revealed ten features (i.e., cortical responses from eight frontocortical fNIRS channels, age and sex) strongly correlated with PTSS severity (r = 0.65, p < .001). Our findings suggest the potential utility of fNIRS as a portable tool for the detection of putative PTSS brain biomarkers.

Sleep disturbance-related neuroimaging features as potential biomarkers for the diagnosis of major depressive disorder: A multicenter study based on machine learning

BACKGROUND

Objective biomarkers are crucial for overcoming the clinical dilemma in major depressive disorder (MDD), and the individualized diagnosis is essential to facilitate the precise medicine for MDD.

METHODS

Sleep disturbance-related magnetic resonance imaging (MRI) features was identified in the internal dataset (92 MDD patients) using the relevance vector regression algorithm, which was further verified in 460 MDD patients of an independent, multicenter dataset. Subsequently, using these MRI features, the eXtreme Gradient Boosting classification model was constructed in the current multicenter dataset (460 MDD patients and 470 normal controls). Meanwhile, the association between classification outputs and the severity of depressive symptoms was also investigated.

RESULTS

In MDD patients, the combination of gray matter density and fractional amplitude of low-frequency fluctuation can accurately predict individual sleep disturbance score that was calculated by the sum of item 4 score, item 5 score, and item 6 score of the 17-Item Hamilton Rating Scale for Depression (HAMD-17) (R² = 0.158 in the internal dataset; R² = 0.110 in multicenter dataset). Furthermore, the classification model based on these MRI features distinguished MDD patients from normal controls with 86.3% accuracy (area under the curve = 0.937). Importantly, the classification outputs significantly correlated with HAMD-17 scores in MDD patients.

LIMITATION

Lacking some specialized tools to assess the personal sleep quality, e.g. Pittsburgh Sleep Quality Index.

CONCLUSION

Neuroimaging features can reflect accurately individual sleep disturbance manifestation and serve as potential diagnostic biomarkers of MDD.

Multivariate morphological brain signatures enable individualized prediction of dispositional need for closure

Need for closure (NFC) reflects stable individual differences in the desire for a quick, definite, and stable answer to a question. A large body of research has documented the association between NFC and various cognitive, emotional and social processes. Despite considerable interest in psychology, little effort has been made to uncover the neural substrates of individual variations in NFC. Herein, we took a data-driven approach to predict NFC trait combining machine learning framework and the whole-brain grey matter volume (GMV) features, which represent a reliable brain imaging measure and have been commonly employed to explore neural basis underlying individual differences of cognition and behaviors. Brain regions contributing to the prediction were then subjected to functional connectivity and decoding analyses for a quantitative inference on their psychophysiological functions. Our results indicated that multivariate patterns of GMV derived from multiple regions across distributed brain systems predicted NFC at individual level. The contributing regions are distributed across the emotional processing network (e.g., striatum), cognitive control network (e.g., lateral prefrontal cortex), social cognition network (e.g., temporoparietal junction) and perceptual processing network (e.g., occipital cortex). The current study provided the first evidence that dispositional NFC is embodied in multiple large-scale brain networks, helping to delineate a more complete picture about the neuropsychological processes that support individual differences in NFC. Beyond these findings, the current interdisciplinary approach to constructing and interpreting neuroimaging-based prediction model of personality traits would be informative to a wide range of future studies on personality.

Identification of the Neural Circuit Underlying Episodic Memory Deficit in Amnestic Mild Cognitive Impairment via Machine Learning on Gray Matter Volume

Based on whole-brain gray matter volume (GMV), we used relevance vector regression to predict the Rey's Auditory Verbal Learning Test Delayed Recall (AVLT-DR) scores of individual amnestic mild cognitive impairment (aMCI) patient. The whole-brain GMV pattern could significantly predict the AVLT-DR scores (r = 0.54, p < 0.001). The most important GMV features mainly involved default-mode (e.g., posterior cingulate gyrus, angular gyrus, and middle temporal gyrus) and limbic systems (e.g., hippocampus and parahippocampal gyrus). Therefore, our results provide evidence supporting the idea that the episodic memory deficit in aMCI patients is associated with disruption of the default-mode and limbic systems.

Deep learning model of fMRI connectivity predicts PTSD symptom trajectories in recent trauma survivors

Early intervention following exposure to a traumatic life event could change the clinical path from the development of post traumatic stress disorder (PTSD) to recovery, hence the interest in early detection and underlying biological mechanisms involved in the development of post traumatic sequelae. We introduce a novel end-to-end neural network that employs resting-state and task-based functional MRI (fMRI) datasets, obtained one month after trauma exposure, to predict PTSD symptoms at one-, six- and fourteen-months after the exposure. FMRI data, as well as PTSD status and symptoms, were collected from adults at risk for PTSD development, after admission to emergency room following a traumatic event. Our computational method utilized a per-region encoder to extract brain regions embedding, which were subsequently updated by applying the algorithmic technique of pairwise attention. The affinities obtained between each pair of regions were combined to create a pairwise co-activation map used to perform multi-label classification. The results demonstrate that the novel method's performance in predicting PTSD symptoms, in a prospective manner, outperforms previous analytical techniques reported in the fMRI literature, all trained on the same dataset. We further show a high predictive ability for predicting PTSD symptom clusters and PTSD persistence. To the best of our knowledge, this is the first deep learning method applied on fMRI data with respect to prospective clinical outcomes, to predict PTSD status, severity and symptom clusters. Future work could further delineate the mechanisms that underlie such a prediction, and potentially improve single patient characterization.

Identification of specific neural circuit underlying the key cognitive deficit of remitted late-onset depression: A multi-modal MRI and machine learning study

Neuropsychological impairment is a key feature of late-onset depression (LOD), with deficits observed across multiple cognitive domains. And this neuropsychological impairment can persist even after the remission of depressive symptoms. However, none of previous studies have explored the pattern of cognitive deficit in remitted LOD (rLOD), and investigated the specific neural circuit underlying the key cognitive deficit of LOD. 40 rLOD patients and 36 controls underwent comprehensive neuropsychological assessments and magnetic resonance imaging (MRI) scans. The influence of executive function or information processing speed deficit on other cognitive domains was first investigated. We then applied a multivariate machine learning technique known as relevance vector regression to evaluate the potential of multiple-modal MRI (i.e., integrating whole-brain grey-matter [GM] volume and white-matter [WM] tract features) for making accurate predictions about the key cognitive deficit for individual rLOD patient. We revealed that the information processing speed appears to represent a key cognitive deficit in rLOD. Further the machine learning model identified a wide range of GM regions and WM tracts that significantly contributed to the prediction of individual performance on information processing speed (r = 0.50, P < 0.001). The GM regions mainly located in the frontal-subcortical and limbic systems; and the WM tracts mainly located in the frontal-limbic pathway, including the anterior corona radiata, fornix, posterior cingulate bundle, and uncinate fasciculus. This present study provide strongly evidence supporting the concept of rLOD that the core aspect of the cognitive deficits (i.e., information processing speed) is associated with disruption of the frontal-subcortical-limbic pathway.

Hippocampal subfield alterations in pediatric patients with post-traumatic stress disorder

The hippocampus, a key structure with distinct subfield functions, is strongly implicated in the pathophysiology of post-traumatic stress disorder (PTSD); however, few studies of hippocampus subfields in PTSD have focused on pediatric patients. We therefore investigated the hippocampal subfield volume using an automated segmentation method and explored the subfield-centered functional connectivity aberrations related to the anatomical changes, in a homogenous population of traumatized children with and without PTSD. To investigate the potential diagnostic value in individual patients, we used a machine learning approach to identify features with significant discriminative power for diagnosis of PTSD using random forest classifiers. Compared to controls, we found significant mean volume reductions of 8.4% and 9.7% in the right presubiculum and hippocampal tail in patients, respectively. These two subfields' volumes were the most significant contributors to group discrimination, with a mean classification accuracy of 69% and a specificity of 81%. These anatomical alterations, along with the altered functional connectivity between (pre)subiculum and inferior frontal gyrus, may underlie deficits in fear circuitry leading to dysfunction of fear extinction and episodic memory, causally important in post-traumatic symptoms such as hypervigilance and re-experience. For the first time, we suggest that hippocampal subfield volumes might be useful in discriminating traumatized children with and without PTSD.

Translational Application of a Neuro-Scientific Multi-Modal Approach Into Forensic Psychiatric Evaluation: Why and How?

A prominent body of literature indicates that insanity evaluations, which are intended to provide influential expert reports for judges to reach a decision "beyond any reasonable doubt," suffer from a low inter-rater reliability. This paper reviews the limitations of the classical approach to insanity evaluation and the criticisms to the introduction of neuro-scientific approach in court. Here, we explain why in our opinion these criticisms, that seriously hamper the translational implementation of neuroscience into the forensic setting, do not survive scientific scrutiny. Moreover, we discuss how the neuro-scientific multimodal approach may improve the inter-rater reliability in insanity evaluation. Critically, neuroscience does not aim to introduce a brain-based concept of insanity. Indeed, criteria for responsibility and insanity are and should remain clinical. Rather, following the falsificationist approach and the convergence of evidence principle, the neuro-scientific multimodal approach is being proposed as a way to improve reliability of insanity evaluation and to mitigate the influence of cognitive biases on the formulation of insanity opinions, with the final aim to reduce errors and controversies.

Individualized Prediction of PTSD Symptom Severity in Trauma Survivors From Whole-Brain Resting-State Functional Connectivity

Previous studies have demonstrated relations between spontaneous neural activity evaluated by resting-state functional magnetic resonance imaging (fMRI) and symptom severity in post-traumatic stress disorder. However, few studies have used brain-based measures to identify imaging associations with illness severity at the level of individual patients. This study applied connectome-based predictive modeling (CPM), a recently developed data-driven and subject-level method, to identify brain function features that are related to symptom severity of trauma survivors. Resting-state fMRI scans and clinical ratings were obtained 10-15 months after the earthquake from 122 earthquake survivors. Symptom severity of post-traumatic stress disorder features for each survivor was evaluated using the Clinician Administered Post-traumatic Stress Disorder Scale (CAPS-IV). A functionally pre-defined atlas was applied to divide the human brain into 268 regions. Each individual's functional connectivity 268 × 268 matrix was created to reflect correlations of functional time series data across each pair of nodes. The relationship between CAPS-IV scores and brain functional connectivity was explored in a CPM linear model. Using a leave-one-out cross-validation (LOOCV) procedure, findings showed that the positive network model predicted the left-out individual's CAPS-IV scores from resting-state functional connectivity. CPM predicted CAPS-IV scores, as indicated by a significant correspondence between predicted and actual values (r = 0.30, P = 0.001) utilizing primarily functional connectivity between visual cortex, subcortical-cerebellum, limbic, and motor systems. The current study provides data-driven evidence regarding the functional brain features that predict symptom severity based on the organization of intrinsic brain networks and highlights its potential application in making clinical evaluation of symptom severity at the individual level.

Multi-voxel pattern analysis of amygdala functional connectivity at rest predicts variability in posttraumatic stress severity

INTRODUCTION

Resting state functional magnetic resonance imaging (rsfMRI) studies demonstrate that individuals with posttraumatic stress disorder (PTSD) exhibit atypical functional connectivity (FC) between the amygdala, involved in the generation of emotion, and regions responsible for emotional appraisal (e.g., insula, orbitofrontal cortex [OFC]) and regulation (prefrontal cortex [PFC], anterior cingulate cortex). Consequently, atypical amygdala FC within an emotional processing and regulation network may be a defining feature of PTSD, although altered FC does not seem constrained to one brain region. Instead, altered amygdala FC involves a large, distributed brain network in those with PTSD. The present study used a machine-learning data-driven approach, multi-voxel pattern analysis (MVPA), to predict PTSD severity based on whole-brain patterns of amygdala FC.

METHODS

Trauma-exposed adults (N = 90) completed the PTSD Checklist-Civilian Version to assess symptoms and a 5-min rsfMRI. Whole-brain FC values to bilateral amygdala were extracted and used in a relevance vector regression analysis with a leave-one-out approach for cross-validation with permutation testing (1,000) to obtain significance values.

RESULTS

Results demonstrated that amygdala FC predicted PCL-C scores with statistically significant accuracy (r = .46, p = .001; mean sum of squares = 130.46, p = .001; R²  = 0.21, p = .001). Prediction was based on whole-brain amygdala FC, although regions that informed prediction (top 10%) included the OFC, amygdala, and dorsolateral PFC.

CONCLUSION

Findings demonstrate the utility of MVPA based on amygdala FC to predict individual severity of PTSD symptoms and that amygdala FC within a fear acquisition and regulation network contributed to accurate prediction.

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.

Diagnostic and Predictive Neuroimaging Biomarkers for Posttraumatic Stress Disorder

BACKGROUND

Comorbidity between posttraumatic stress disorder (PTSD) and major depressive disorder (MDD) has been commonly overlooked by studies examining resting-state functional connectivity patterns in PTSD. The current study used a data-driven approach to identify resting-state functional connectivity biomarkers to 1) differentiate individuals with PTSD (with or without MDD) from trauma-exposed healthy control subjects (TEHCs), 2) compare individuals with PTSD alone with those with comorbid PTSD+MDD, and 3) explore the clinical utility of the identified biomarkers by testing their associations with clinical symptoms and treatment response.

METHODS

Resting-state magnetic resonance images were obtained from 51 individuals with PTSD alone, 52 individuals with PTSD+MDD, and 76 TEHCs. Of the 103 individuals with PTSD, 55 were enrolled in prolonged exposure treatment. A support vector machine model was used to identify resting-state functional connectivity biomarkers differentiating individuals with PTSD (with or without MDD) from TEHCs and differentiating individuals with PTSD alone from those with PTSD+MDD. The associations between the identified features and symptomatology were tested with Pearson correlations.

RESULTS

The support vector machine model achieved 70.6% accuracy in discriminating between individuals with PTSD and TEHCs and achieved 76.7% accuracy in discriminating between individuals with PTSD alone and those with PTSD+MDD for out-of-sample prediction. Within-network connectivity in the executive control network, prefrontal network, and salience network discriminated individuals with PTSD from TEHCs. The basal ganglia network played an important role in differentiating individuals with PTSD alone from those with PTSD+MDD. PTSD scores were inversely correlated with within-executive control network connectivity (p < .001), and executive control network connectivity was positively correlated with treatment response (p < .001).

CONCLUSIONS

Results suggest that unique brain-based abnormalities differentiate individuals with PTSD from TEHCs, differentiate individuals with PTSD from those with PTSD+MDD, and demonstrate clinical utility in predicting levels of symptomatology and treatment response.

Hippocampal and parahippocampal volumes vary by sex and traumatic life events in children

BACKGROUND

Childhood trauma is reliably associated with smaller hippocampal volume in adults; however, this finding has not been shown in children, and even less is known about how sex and trauma interact to affect limbic structural development in children.

METHODS

Typically developing children aged 9 to 15 years who completed a trauma history questionnaire and structural T1-weighted MRI were included in this study (n = 172; 85 female, 87 male). All children who reported 4 or more traumas (n = 36) composed the high trauma group, and all children who reported 3 or fewer traumas (n = 136) composed the low trauma group. Using multivariate analysis of covariance, we compared FreeSurfer-derived structural MRI volumes (normalized by total intracranial volume) of the amygdalar, hippocampal and parahippocampal regions by sex and trauma level, controlling for age and study site.

RESULTS

We found a significant sex × trauma interaction, such that girls with high trauma had greater volumes than boys with high trauma. Follow-up analyses indicated significantly increased volumes for girls and generally decreased volumes for boys, specifically in the hippocampal and parahippocampalregions for the high trauma group; we observed no sex differences in the low trauma group. We noted no interaction effect for the amygdalae.

LIMITATIONS

We assessed a community sample and did not include a clinical sample. We did not collect data about the ages at which children experienced trauma.

CONCLUSION

Results revealed that psychological trauma affects brain development differently in girls and boys. These findings need to be followed longitudinally to elucidate how structural differences progress and contribute to well-known sex disparities in psychopathology.

Classifying heterogeneous presentations of PTSD via the default mode, central executive, and salience networks with machine learning

Intrinsic connectivity networks (ICNs), including the default mode network (DMN), the central executive network (CEN), and the salience network (SN) have been shown to be aberrant in patients with posttraumatic stress disorder (PTSD). The purpose of the current study was to a) compare ICN functional connectivity between PTSD, dissociative subtype PTSD (PTSD+DS) and healthy individuals; and b) to examine the use of multivariate machine learning algorithms in classifying PTSD, PTSD+DS, and healthy individuals based on ICN functional activation. Our neuroimaging dataset consisted of resting-state fMRI scans from 186 participants [PTSD (n = 81); PTSD + DS (n = 49); and healthy controls (n = 56)]. We performed group-level independent component analyses to evaluate functional connectivity differences within each ICN. Multiclass Gaussian Process Classification algorithms within PRoNTo software were then used to predict the diagnosis of PTSD, PTSD+DS, and healthy individuals based on ICN functional activation. When comparing the functional connectivity of ICNs between PTSD, PTSD+DS and healthy controls, we found differential patterns of connectivity to brain regions involved in emotion regulation, in addition to limbic structures and areas involved in self-referential processing, interoception, bodily self-consciousness, and depersonalization/derealization. Machine learning algorithms were able to predict with high accuracy the classification of PTSD, PTSD+DS, and healthy individuals based on ICN functional activation. Our results suggest that alterations within intrinsic connectivity networks may underlie unique psychopathology and symptom presentation among PTSD subtypes. Furthermore, the current findings substantiate the use of machine learning algorithms for classifying subtypes of PTSD illness based on ICNs.

Acute Posttrauma Resting-State Functional Connectivity of Periaqueductal Gray Prospectively Predicts Posttraumatic Stress Disorder Symptoms

BACKGROUND

Posttraumatic stress disorder (PTSD) is characterized by hyperarousal, avoidance, and intrusive/re-experiencing symptoms. The periaqueductal gray (PAG), which generates behavioral responses to physical and psychological stressors, is also implicated in threat processing. Distinct regions of the PAG elicit opposing responses to threatening or stressful stimuli; the ventrolateral PAG evokes passive coping strategies (e.g., analgesia), whereas the dorsolateral PAG (dlPAG) promotes active responses (e.g., fight or flight). We investigated whether altered PAG resting-state functional connectivity (RSFC) prospectively predicted PTSD symptoms.

METHODS

A total of 48 trauma-exposed individuals underwent an RSFC scan 2 weeks posttraumatic injury. Self-report measures, including the visual analog scale for pain and the Impact of Event Scale, were collected at 2 weeks and 6 months posttrauma. We analyzed whether acute bilateral PAG RSFC was a marker of risk for total 6-month symptom severity and specific symptom clusters. In an exploratory analysis, we investigated whether dlPAG RSFC predicted PTSD symptoms.

RESULTS

After adjusting for physical pain ratings, greater acute posttrauma PAG-frontal pole and PAG-posterior cingulate cortex connectivity was positively associated with 6-month total PTSD symptoms. Weaker dlPAG-superior/inferior parietal lobule connectivity predicted both higher hyperarousal and higher intrusive symptoms, while weaker dlPAG-supramarginal gyrus RSFC was associated with only hyperarousal symptoms.

CONCLUSIONS

Altered connectivity of the PAG 2 weeks posttrauma prospectively predicted PTSD symptoms. These findings suggest that aberrant PAG function may serve as a marker of risk for chronic PTSD symptoms, possibly by driving specific symptom clusters, and more broadly that connectivity of specific brain regions may underlie specific symptom profiles.

Uncovering the heterogeneity of posttraumatic stress disorder: Towards a personalized medicine approach for military members and Veterans

Introduction: Recently, there has been substantial interest in exploring the heterogeneity of posttraumatic stress disorder (PTSD) on a neurobiological level, as individuals with PTSD, including military members and Veterans, vary in their presentation of symptoms. Methods: Critically, a dissociative subtype of PTSD (PTSD+DS) has been defined, where a large body of evidence suggests that the unique presentation of symptoms among PTSD+DS patients is associated with aberrant neurobiological underpinnings. Results: PTSD+DS is often characterized by emotion overmodulation, with increased top-down activation from emotion regulation areas, which is associated with emotional detachment, depersonalization, and derealization. This is in stark contrast to the symptoms commonly observed in individuals with PTSD, who exhibit emotion undermodulation, which involves decreased top-down regulation of hyperactive emotion generation areas and is associated with vivid re-experiencing of trauma memories and hyperarousal. Discussion: This article examines a clinical case example that clearly illustrates this heterogeneous presentation of PTSD symptomatology and psychopathology. It discusses the implications this evidence base holds for a neurobiologically-informed, personalized medicine approach to treatment for military members and Veterans.

Exploring memory function in earthquake trauma survivors with resting-state fMRI and machine learning

BACKGROUND

Traumatized earthquake survivors may develop poor memory function. Resting-state functional magnetic resonance imaging (rs-fMRI) and machine learning techniques may one day aid the clinical assessment of individual psychiatric patients. This study aims to use machine learning with Rs-fMRI from the perspectives of neurophysiology and neuroimaging to explore the association between it and the individual memory function of trauma survivors.

METHODS

Rs-fMRI data was acquired for eighty-nine survivors (male (33%), average age (SD):45.18(6.31) years) of Wenchuan earthquakes in 2008 each of whom was screened by experienced psychiatrists based on the clinician-administered post-traumatic stress disorder (PTSD) scale (CAPS), and their memory function scores were determined by the Wechsler Memory Scale-IV (WMS-IV). We explored which memory function scores were significantly associated with CAPS scores. Using simple multiple kernel learning (MKL), Rs-fMRI was used to predict the memory function scores that were associated with CAPS scores. A support vector machine (SVM) was also used to make classifications in trauma survivors with or without PTSD.

RESULTS

Spatial addition (SA), which is defined by spatial working memory function, was negatively correlated with the total CAPS score (r = - 0.22, P = 0.04). The use of simple MKL allowed quantitative association of SA scores with statistically significant accuracy (correlation = 0.28, P = 0.03; mean squared error = 8.36; P = 0.04). The left middle frontal gyrus and the left precuneus contributed the largest proportion to the simple MKL association frame. The SVM could not make a quantitative classification of diagnosis with statistically significant accuracy.

LIMITATIONS

The use of the cross-sectional study design after exposure to an earthquake and the leave-one-out cross-validation (LOOCV) increases the risk of overfitting.

CONCLUSION

Spontaneous brain activity of the left middle frontal gyrus and the left precuneus acquired by rs-fMRI may be a brain mechanism of visual working memory that is related to PTSD symptoms. Machine learning may be a useful tool in the identification of brain mechanisms of memory impairment in trauma survivors.

Intrinsic sensory disinhibition contributes to intrusive re-experiencing in combat veterans

Intrusive re-experiencing of traumatic events is a hallmark symptom of posttraumatic stress disorder, characterized by rich and vivid sensory details as reported in "flashbacks". While prevailing models of trauma intrusions focus on dysregulated emotional processes, we hypothesize that a deficiency in intrinsic sensory inhibition could drive overactivation of sensory representations of trauma memories, precipitating sensory-rich intrusions. In a sample of combat veterans, we examined resting-state alpha (8-12 Hz) oscillatory activity (in both power and posterior→frontal connectivity), given its role in sensory cortical inhibition, in association with intrusive re-experiencing symptoms. Veterans further participated in an odor task (including both combat and non-combat odors) to assess olfactory trauma memory and emotional response. We observed an association between intrusive re-experiencing symptoms and attenuated resting-state posterior→frontal alpha connectivity, which were both correlated with olfactory trauma memory. Importantly, olfactory trauma memory was identified as a mediator of the relationship between alpha connectivity and intrusive re-experiencing, suggesting that deficits in intrinsic sensory inhibition contributed to intrusive re-experiencing of trauma via heightened trauma memory. Therefore, by permitting unfiltered sensory cues to enter information processing and activate sensory representations of trauma, sensory disinhibition can constitute a sensory mechanism of intrusive re-experiencing in trauma-exposed individuals.

Differences in functional connectivity density among subtypes of schizophrenic auditory hallucination

This study aimed to investigate alterations in brain function among different subtypes of auditory hallucinations (AH) in drug-naïve first episode schizophrenia patients. We recruited 20 patients with drug-naïve first episode schizophrenia who had constant commanding and commenting auditory verbal hallucinations (CCCAVH), 15 drug-naïve first episode schizophrenia patients who had nonverbal auditory hallucinations (NVAH), and 20 healthy controls to participate in this study. We used global functional connectivity density (gFCD) and one-way analysis of covariance to characterize differences in brain function between the two patient groups. Statistical significance was set at P < 0.05. As compared to controls, schizophrenia patients with CCCAVH demonstrated increased gFCD in the right Broca's area, bilateral superior temporal gyri, hippocampus, bilateral insula, and anterior cingulate gyri, and decreased gFCD in the left temporoparietal junction (family-wise error [FEW] correct, P < 0.05). Schizophrenia patients with NVAH demonstrated increased gFCD in the bilateral superior temporal gyri and most of the components of the default mode network (DMN), and decreased gFCD in components of the executive control network (FWE correct, P < 0.05). We found that schizophrenia patients with CCCAVH and NVAH have distinct functional brain patterns. The features observed in patients with CCCAVH are consistent with the "inner speech" hypothesis of AH. Features of patients with NVAH suggest hyperactivity of the superior temporal gyrus and DMN, and hypoactivity of the prefrontal lobe.

Machine learning in resting-state fMRI analysis

Machine learning techniques have gained prominence for the analysis of resting-state functional Magnetic Resonance Imaging (rs-fMRI) data. Here, we present an overview of various unsupervised and supervised machine learning applications to rs-fMRI. We offer a methodical taxonomy of machine learning methods in resting-state fMRI. We identify three major divisions of unsupervised learning methods with regard to their applications to rs-fMRI, based on whether they discover principal modes of variation across space, time or population. Next, we survey the algorithms and rs-fMRI feature representations that have driven the success of supervised subject-level predictions. The goal is to provide a high-level overview of the burgeoning field of rs-fMRI from the perspective of machine learning applications.

Partial Least Squares Regression Performs Well in MRI-Based Individualized Estimations

Estimation of individuals' cognitive, behavioral and demographic (CBD) variables based on MRI has attracted much research interest in the past decade, and effective machine learning techniques are of great importance for these estimations. Partial least squares regression (PLSR) is an attractive machine learning technique that can accommodate both single- and multi-label learning in a simple framework, while its potential for MRI-based estimations of CBD variables remains to be explored. In this study, we systemically investigated the performance of PLSR in MRI-based estimations of individuals' CBD variables, especially its performance in simultaneous estimation of multiple CBD variables (multi-label learning). We performed the study on the dataset included in the HCP S1200 release. Resting state functional connections (RSFCs) were used as features, and a total of 10 CBD variables (e.g., age, gender, grip strength, and picture vocabulary) were estimated. The results showed that PLSR performed well in both single- and multi-label learning. In fact, the present estimations were better than those reported in literatures, as indicated by stronger correlations between the estimated and actual CBD variables, as well as high gender classification accuracy (97.8% in this study). Moreover, the RSFCs that contributed to the estimations exhibited strong correlations with the CBD variable estimated, that is, PLSR algorithm automatically selected the RSFCs closely related to one CBD variable to establish predictive models for the variable. Besides, the estimation accuracies based on RSFCs among 100, 200, and 300 regions of interest (ROIs) were higher than those based on RSFCs among 15, 25, and 50 ROIs; the estimation accuracies based on RSFCs evaluated using partial correlation were higher than those based on RSFCs evaluated using full correlation. In addition to the aforementioned virtues, PLSR is efficient in model training and testing, and it is simple and easy to use. Therefore, PLSR can be a favorable choice for future MRI-based estimations of CBD variables.

Individualized prediction of dispositional worry using white matter connectivity

BACKGROUND

Excessive worry is a defining feature of generalized anxiety disorder and is present in a wide range of other psychiatric conditions. Therefore, individualized predictions of worry propensity could be highly relevant in clinical practice, with respect to the assessment of worry symptom severity at the individual level.

METHODS

We applied a multivariate machine learning approach to predict dispositional worry based on microstructural integrity of white matter (WM) tracts.

RESULTS

We demonstrated that the machine learning model was able to decode individual dispositional worry scores from microstructural properties in widely distributed WM tracts (mean absolute error = 10.46, p < 0.001; root mean squared error = 12.82, p < 0.001; prediction R2 = 0.17, p < 0.001). WM tracts that contributed to worry prediction included the posterior limb of internal capsule, anterior corona radiate, and cerebral peduncle, as well as the corticolimbic pathways (e.g. uncinate fasciculus, cingulum, and fornix) already known to be critical for emotion processing and regulation.

CONCLUSIONS

The current work thus elucidates potential neuromarkers for clinical assessment of worry symptoms across a wide range of psychiatric disorders. In addition, the identification of widely distributed pathways underlying worry propensity serves to better improve the understanding of the neurobiological mechanisms associated with worry.

Machine learning multivariate pattern analysis predicts classification of posttraumatic stress disorder and its dissociative subtype: a multimodal neuroimaging approach

BACKGROUND

The field of psychiatry would benefit significantly from developing objective biomarkers that could facilitate the early identification of heterogeneous subtypes of illness. Critically, although machine learning pattern recognition methods have been applied recently to predict many psychiatric disorders, these techniques have not been utilized to predict subtypes of posttraumatic stress disorder (PTSD), including the dissociative subtype of PTSD (PTSD + DS).

METHODS

Using Multiclass Gaussian Process Classification within PRoNTo, we examined the classification accuracy of: (i) the mean amplitude of low-frequency fluctuations (mALFF; reflecting spontaneous neural activity during rest); and (ii) seed-based amygdala complex functional connectivity within 181 participants [PTSD (n = 81); PTSD + DS (n = 49); and age-matched healthy trauma-unexposed controls (n = 51)]. We also computed mass-univariate analyses in order to observe regional group differences [false-discovery-rate (FDR)-cluster corrected p < 0.05, k = 20].

RESULTS

We found that extracted features could predict accurately the classification of PTSD, PTSD + DS, and healthy controls, using both resting-state mALFF (91.63% balanced accuracy, p < 0.001) and amygdala complex connectivity maps (85.00% balanced accuracy, p < 0.001). These results were replicated using independent machine learning algorithms/cross-validation procedures. Moreover, areas weighted as being most important for group classification also displayed significant group differences at the univariate level. Here, whereas the PTSD + DS group displayed increased activation within emotion regulation regions, the PTSD group showed increased activation within the amygdala, globus pallidus, and motor/somatosensory regions.

CONCLUSION

The current study has significant implications for advancing machine learning applications within the field of psychiatry, as well as for developing objective biomarkers indicative of diagnostic heterogeneity.

Connectome-based model predicts individual differences in propensity to trust

Trust constitutes a fundamental basis of human society and plays a pivotal role in almost every aspect of human relationships. Although enormous interest exists in determining the neuropsychological underpinnings of a person's propensity to trust utilizing task-based fMRI; however, little progress has been made in predicting its variations by task-free fMRI based on whole-brain resting-state functional connectivity (RSFC). Here, we combined a one-shot trust game with a connectome-based predictive modeling approach to predict propensity to trust from whole-brain RSFC. We demonstrated that individual variations in the propensity to trust were primarily predicted by RSFC rooted in the functional integration of distributed key nodes-caudate, amygdala, lateral prefrontal cortex, temporal-parietal junction, and the temporal pole-which are part of domain-general large-scale networks essential for the motivational, affective, and cognitive aspects of trust. We showed, further, that the identified brain-behavior associations were only evident for trust but not altruistic preferences and that propensity to trust (and its underlying neural underpinnings) were modulated according to the extent to which a person emphasizes general social preferences (i.e., horizontal collectivism) rather than general risk preferences (i.e., trait impulsiveness). In conclusion, the employed data-driven approach enables to predict propensity to trust from RSFC and highlights its potential use as an objective neuromarker of trust impairment in mental disorders.

EEG Microstates Temporal Dynamics Differentiate Individuals with Mood and Anxiety Disorders From Healthy Subjects

Electroencephalography (EEG) measures the brain’s electrophysiological spatio-temporal activities with high temporal resolution. Multichannel and broadband analysis of EEG signals is referred to as EEG microstates (EEG-ms) and can characterize such dynamic neuronal activity. EEG-ms have gained much attention due to the increasing evidence of their association with mental activities and large-scale brain networks identified by functional magnetic resonance imaging (fMRI). Spatially independent EEG-ms are quasi-stationary topographies (e.g., stable, lasting a few dozen milliseconds) typically classified into four canonical classes (microstates A through D). They can be identified by clustering EEG signals around EEG global field power (GFP) maxima points. We examined the EEG-ms properties and the dynamics of cohorts of mood and anxiety (MA) disorders subjects (n = 61) and healthy controls (HCs; n = 52). In both groups, we found four distinct classes of EEG-ms (A through D), which did not differ among cohorts. This suggests a lack of significant structural cortical abnormalities among cohorts, which would otherwise affect the EEG-ms topographies. However, both cohorts’ brain network dynamics significantly varied, as reflected in EEG-ms properties. Compared to HC, the MA cohort features a lower transition probability between EEG-ms B and D and higher transition probability from A to D and from B to C, with a trend towards significance in the average duration of microstate C. Furthermore, we harnessed a recently introduced theoretical approach to analyze the temporal dependencies in EEG-ms. The results revealed that the transition matrices of MA group exhibit higher symmetrical and stationarity properties as compared to HC ones. In addition, we found an elevation in the temporal dependencies among microstates, especially in microstate B for the MA group. The determined alteration in EEG-ms temporal dependencies among the cohorts suggests that brain abnormalities in mood and anxiety disorders reflect aberrant neural dynamics and a temporal dwelling among ceratin brain states (i.e., mood and anxiety disorders subjects have a less dynamicity in switching between different brain states).

Resting-state Functional Connectivity and Deception: Exploring Individualized Deceptive Propensity by Machine Learning

Individuals show marked variability in determining to be honest or deceptive in daily life. A large number of studies have investigated the neural substrates of deception; however, the brain networks contributing to the individual differences in deception remain unclear. In this study, we sought to address this issue by employing a machine-learning approach to predict individuals' deceptive propensity based on the topological properties of whole-brain resting-state functional connectivity (RSFC). Participants finished the resting-state functional MRI (fMRI) data acquisition, and then, one week later, participated as proposers in a modified ultimatum game in which they spontaneously chose to be honest or deceptive. A linear relevance vector regression (RVR) model was trained and validated to examine the relationship between topological properties of networks of RSFC and actual deceptive behaviors. The machine-learning model sufficiently decoded individual differences in deception using three brain networks based on RSFC, including the executive controlling network (dorsolateral prefrontal cortex, middle frontal cortex, and orbitofrontal cortex), the social and mentalizing network (the temporal lobe, temporo-parietal junction, and inferior parietal lobule), and the reward network (putamen and thalamus). These networks have been found to form a signaling cognitive framework of deception by coding the mental states of others and the reward or values of deception or honesty, and integrating this information to make a final decision about being deceptive or honest. These findings suggest the potential of using RSFC as a task-independent neural trait for predicting deceptive propensity, and shed light on using machine-learning approaches in deception detection.

Associations between polygenic risk scores for four psychiatric illnesses and brain structure using multivariate pattern recognition

Psychiatric illnesses are complex and polygenic. They are associated with widespread alterations in the brain, which are partly influenced by genetic factors. There have been some attempts to relate polygenic risk scores (PRS) - a measure of the overall genetic risk an individual carries for a disorder - to brain structure using univariate methods. However, PRS are likely associated with distributed and covarying effects across the brain. We therefore used multivariate machine learning in this proof-of-principle study to investigate associations between brain structure and PRS for four psychiatric disorders; attention deficit-hyperactivity disorder (ADHD), autism, bipolar disorder and schizophrenia. The sample included 213 individuals comprising patients with depression (69), bipolar disorder (33), and healthy controls (111). The five psychiatric PRSs were calculated based on summary data from the Psychiatric Genomics Consortium. T1-weighted magnetic resonance images were obtained and voxel-based morphometry was implemented in SPM12. Multivariate relevance vector regression was implemented in the Pattern Recognition for Neuroimaging Toolbox (PRoNTo). Across the whole sample, a multivariate pattern of grey matter significantly predicted the PRS for autism (r = 0.20, pFDR = 0.03; MSE = 4.20 × 10-5, pFDR = 0.02). For the schizophrenia PRS, the MSE was significant (MSE = 1.30 × 10-5, pFDR = 0.02) although the correlation was not (r = 0.15, pFDR = 0.06). These results lend support to the hypothesis that polygenic liability for autism and schizophrenia is associated with widespread changes in grey matter concentrations. These associations were seen in individuals not affected by these disorders, indicating that this is not driven by the expression of the disease, but by the genetic risk captured by the PRSs.

The effect of machine learning regression algorithms and sample size on individualized behavioral prediction with functional connectivity features

Individualized behavioral/cognitive prediction using machine learning (ML) regression approaches is becoming increasingly applied. The specific ML regression algorithm and sample size are two key factors that non-trivially influence prediction accuracies. However, the effects of the ML regression algorithm and sample size on individualized behavioral/cognitive prediction performance have not been comprehensively assessed. To address this issue, the present study included six commonly used ML regression algorithms: ordinary least squares (OLS) regression, least absolute shrinkage and selection operator (LASSO) regression, ridge regression, elastic-net regression, linear support vector regression (LSVR), and relevance vector regression (RVR), to perform specific behavioral/cognitive predictions based on different sample sizes. Specifically, the publicly available resting-state functional MRI (rs-fMRI) dataset from the Human Connectome Project (HCP) was used, and whole-brain resting-state functional connectivity (rsFC) or rsFC strength (rsFCS) were extracted as prediction features. Twenty-five sample sizes (ranged from 20 to 700) were studied by sub-sampling from the entire HCP cohort. The analyses showed that rsFC-based LASSO regression performed remarkably worse than the other algorithms, and rsFCS-based OLS regression performed markedly worse than the other algorithms. Regardless of the algorithm and feature type, both the prediction accuracy and its stability exponentially increased with increasing sample size. The specific patterns of the observed algorithm and sample size effects were well replicated in the prediction using re-testing fMRI data, data processed by different imaging preprocessing schemes, and different behavioral/cognitive scores, thus indicating excellent robustness/generalization of the effects. The current findings provide critical insight into how the selected ML regression algorithm and sample size influence individualized predictions of behavior/cognition and offer important guidance for choosing the ML regression algorithm or sample size in relevant investigations.

Individualized prediction of trait narcissism from whole-brain resting-state functional connectivity

Narcissism is one of the most fundamental personality traits in which individuals in general population exhibit a large heterogeneity. Despite a surge of interest in examining behavioral characteristics of narcissism in the past decades, the neurobiological substrates underlying narcissism remain poorly understood. Here, we addressed this issue by applying a machine learning approach to decode trait narcissism from whole-brain resting-state functional connectivity (RSFC). Resting-state functional MRI (fMRI) data were acquired for a large sample comprising 155 healthy adults, each of whom was assessed for trait narcissism. Using a linear prediction model, we examined the relationship between whole-brain RSFC and trait narcissism. We demonstrated that the machine-learning model was able to decode individual trait narcissism from RSFC across multiple neural systems, including functional connectivity between and within limbic and prefrontal systems as well as their connectivity with other networks. Key nodes that contributed to the prediction model included the amygdala, prefrontal and anterior cingulate regions that have been linked to trait narcissism. These findings remained robust using different validation procedures. Our findings thus demonstrate that RSFC among multiple neural systems predicts trait narcissism at the individual level.

Brain structural covariance network centrality in maltreated youth with PTSD and in maltreated youth resilient to PTSD

Child maltreatment is a major cause of pediatric posttraumatic stress disorder (PTSD). Previous studies have not investigated potential differences in network architecture in maltreated youth with PTSD and those resilient to PTSD. High-resolution magnetic resonance imaging brain scans at 3 T were completed in maltreated youth with PTSD (n = 31), without PTSD (n = 32), and nonmaltreated controls (n = 57). Structural covariance network architecture was derived from between-subject intraregional correlations in measures of cortical thickness in 148 cortical regions (nodes). Interregional positive partial correlations controlling for demographic variables were assessed, and those correlations that exceeded specified thresholds constituted connections in cortical brain networks. Four measures of network centrality characterized topology, and the importance of cortical regions (nodes) within the network architecture were calculated for each group. Permutation testing and principle component analysis method were employed to calculate between-group differences. Principle component analysis is a methodological improvement to methods used in previous brain structural covariance network studies. Differences in centrality were observed between groups. Larger centrality was found in maltreated youth with PTSD in the right posterior cingulate cortex; smaller centrality was detected in the right inferior frontal cortex compared to youth resilient to PTSD and controls, demonstrating network characteristics unique to pediatric maltreatment-related PTSD. Larger centrality was detected in right frontal pole in maltreated youth resilient to PTSD compared to youth with PTSD and controls, demonstrating structural covariance network differences in youth resilience to PTSD following maltreatment. Smaller centrality was found in the left posterior cingulate cortex and in the right inferior frontal cortex in maltreated youth compared to controls, demonstrating attributes of structural covariance network topology that is unique to experiencing maltreatment. This work is the first to identify cortical thickness-based structural covariance network differences between maltreated youth with and without PTSD. We demonstrated network differences in both networks unique to maltreated youth with PTSD and those resilient to PTSD. The networks identified are important for the successful attainment of age-appropriate social cognition, attention, emotional processing, and inhibitory control. Our findings in maltreated youth with PTSD versus those without PTSD suggest vulnerability mechanisms for developing PTSD.

Attention training modulates resting-state neurophysiological abnormalities in posttraumatic stress disorder

Recent research indicates the relative benefits of computerized attention control treatment (ACT) and attention bias modification treatment (ABMT) for posttraumatic stress disorder (PTSD); however, neural changes underlying these therapeutic effects remain unknown. This study examines how these two types of attention training modulate neurological dysfunction in veterans with PTSD. A community sample of 46 combat veterans with PTSD participated in a randomized double-blinded clinical trial of ACT versus ABMT and 32 of those veterans also agreed to undergo resting-state magnetoencephalography (MEG) recordings. Twenty-four veterans completed psychological and MEG assessments at pre- and post-training to evaluate treatment effects. MEG data were imaged using an advanced Bayesian reconstruction method and examined using statistical parametric mapping. In this report, we focus on the neural correlates and the differential treatment effects observed using MEG; the results of the full clinical trial have been described elsewhere. Our results indicated that ACT modulated occipital and ABMT modulated medial temporal activity more strongly than the comparative treatment. PTSD symptoms decreased significantly from pre- to post-test. These initial neurophysiological outcome data suggest that ACT modulates visual pathways, while ABMT modulates threat-processing regions, but that both are associated with normalizing aberrant neural activity in veterans with PTSD.

Resting-state functional connectivity remains unaffected by preceding exposure to aversive visual stimuli

While much is known about immediate brain activity changes induced by the confrontation with emotional stimuli, the subsequent temporal unfolding of emotions has yet to be explored. To investigate whether exposure to emotionally aversive pictures affects subsequent resting-state networks differently from exposure to neutral pictures, a resting-state fMRI study implementing a two-group repeated-measures design in healthy young adults (N = 34) was conducted. We focused on investigating (i) patterns of amygdala whole-brain and hippocampus connectivity in both a seed-to-voxel and seed-to-seed approach, (ii) whole-brain resting-state networks with an independent component analysis coupled with dual regression, and (iii) the amygdala's fractional amplitude of low frequency fluctuations, all while EEG recording potential fluctuations in vigilance. In spite of the successful emotion induction, as demonstrated by stimuli rating and a memory-facilitating effect of negative emotionality, none of the resting-state measures was differentially affected by picture valence. In conclusion, resting-state networks connectivity as well as the amygdala's low frequency oscillations appear to be unaffected by preceding exposure to widely used emotionally aversive visual stimuli in healthy young adults.

Psychoradiology: The Frontier of Neuroimaging in Psychiatry

Unlike neurologic conditions, such as brain tumors, dementia, and stroke, the neural mechanisms for all psychiatric disorders remain unclear. A large body of research obtained with structural and functional magnetic resonance imaging, positron emission tomography/single photon emission computed tomography, and optical imaging has demonstrated regional and illness-specific brain changes at the onset of psychiatric disorders and in individuals at risk for such disorders. Many studies have shown that psychiatric medications induce specific measurable changes in brain anatomy and function that are related to clinical outcomes. As a result, a new field of radiology, termed psychoradiology, seems primed to play a major clinical role in guiding diagnostic and treatment planning decisions in patients with psychiatric disorders. This article will present the state of the art in this area, as well as perspectives regarding preparations in the field of radiology for its evolution. Furthermore, this article will (a) give an overview of the imaging and analysis methods for psychoradiology; (b) review the most robust and important radiologic findings and their potential clinical value from studies of major psychiatric disorders, such as depression and schizophrenia; and (c) describe the main challenges and future directions in this field. An ongoing and iterative process of developing biologically based nomenclatures with which to delineate psychiatric disorders and translational research to predict and track response to different therapeutic drugs is laying the foundation for a shift in diagnostic practice in psychiatry from a psychologic symptom-based approach to an imaging-based approach over the next generation. This shift will require considerable innovations for the acquisition, analysis, and interpretation of brain images, all of which will undoubtedly require the active involvement of radiologists. ^© RSNA, 2016 Online supplemental material is available for this article.

Resting-State Neurophysiological Abnormalities in Posttraumatic Stress Disorder: A Magnetoencephalography Study

Posttraumatic stress disorder (PTSD) is a debilitating psychiatric condition that is common in veterans returning from combat operations. While the symptoms of PTSD have been extensively characterized, the neural mechanisms that underlie PTSD are only vaguely understood. In this study, we examined the neurophysiology of PTSD using magnetoencephalography (MEG) in a sample of veterans with and without PTSD. Our primary hypothesis was that veterans with PTSD would exhibit aberrant activity across multiple brain networks, especially those involving medial temporal and frontal regions. To this end, we examined a total of 51 USA combat veterans with a battery of clinical interviews and tests. Thirty-one of the combat veterans met diagnostic criteria for PTSD and the remaining 20 did not have PTSD. All participants then underwent high-density MEG during an eyes-closed resting-state task, and the resulting data were analyzed using a Bayesian image reconstruction method. Our results indicated that veterans with PTSD had significantly stronger neural activity in prefrontal, sensorimotor and temporal areas compared to those without PTSD. Veterans with PTSD also exhibited significantly stronger activity in the bilateral amygdalae, parahippocampal and hippocampal regions. Conversely, healthy veterans had stronger neural activity in the bilateral occipital cortices relative to veterans with PTSD. In conclusion, these data suggest that veterans with PTSD exhibit aberrant neural activation in multiple cortical areas, as well as medial temporal structures implicated in affective processing.

Realising stratified psychiatry using multidimensional signatures and trajectories

Background

Stratified or personalised medicine targets treatments for groups of individuals with a disorder based on individual heterogeneity and shared factors that influence the likelihood of response. Psychiatry has traditionally defined diagnoses by constellations of co-occurring signs and symptoms that are assigned a categorical label (e.g. schizophrenia). Trial methodology in psychiatry has evaluated interventions targeted at these categorical entities, with diagnoses being equated to disorders. Recent insights into both the nosology and neurobiology of psychiatric disorder reveal that traditional categorical diagnoses cannot be equated with disorders. We argue that current quantitative methodology (1) inherits these categorical assumptions, (2) allows only for the discovery of average treatment response, (3) relies on composite outcome measures and (4) sacrifices valuable predictive information for stratified and personalised treatment in psychiatry.

Methods and findings

To achieve a truly ‘stratified psychiatry’ we propose and then operationalise two necessary steps: first, a formal multi-dimensional representation of disorder definition and clinical state, and second, the similar redefinition of outcomes as multidimensional constructs that can expose within- and between-patient differences in response. We use the categorical diagnosis of schizophrenia—conceptualised as a label for heterogeneous disorders—as a means of introducing operational definitions of stratified psychiatry using principles from multivariate analysis. We demonstrate this framework by application to the Clinical Antipsychotic Trials of Intervention Effectiveness dataset, showing heterogeneity in both patient clinical states and their trajectories after treatment that are lost in the traditional categorical approach with composite outcomes. We then systematically review a decade of registered clinical trials for cognitive deficits in schizophrenia highlighting existing assumptions of categorical diagnoses and aggregate outcomes while identifying a small number of trials that could be reanalysed using our proposal.

Conclusion

We describe quantitative methods for the development of a multi-dimensional model of clinical state, disorders and trajectories which practically realises stratified psychiatry. We highlight the potential for recovering existing trial data, the implications for stratified psychiatry in trial design and clinical treatment and finally, describe different kinds of probabilistic reasoning tools necessary to implement stratification.

Using deep learning to investigate the neuroimaging correlates of psychiatric and neurological disorders: Methods and applications

Deep learning (DL) is a family of machine learning methods that has gained considerable attention in the scientific community, breaking benchmark records in areas such as speech and visual recognition. DL differs from conventional machine learning methods by virtue of its ability to learn the optimal representation from the raw data through consecutive nonlinear transformations, achieving increasingly higher levels of abstraction and complexity. Given its ability to detect abstract and complex patterns, DL has been applied in neuroimaging studies of psychiatric and neurological disorders, which are characterised by subtle and diffuse alterations. Here we introduce the underlying concepts of DL and review studies that have used this approach to classify brain-based disorders. The results of these studies indicate that DL could be a powerful tool in the current search for biomarkers of psychiatric and neurologic disease. We conclude our review by discussing the main promises and challenges of using DL to elucidate brain-based disorders, as well as possible directions for future research.