Using deep learning to classify pediatric posttraumatic stress disorder at the individual level
Ontology highlight
ABSTRACT: Background
Children exposed to natural disasters are vulnerable to developing posttraumatic stress disorder (PTSD). Previous studies using resting-state functional neuroimaging have revealed alterations in graph-based brain topological network metrics in pediatric PTSD patients relative to healthy controls (HC). Here we aimed to apply deep learning (DL) models to neuroimaging markers of classification which may be of assistance in diagnosis of pediatric PTSD. Methods
We studied 33 pediatric PTSD and 53 matched HC. Functional connectivity between 90 brain regions from the automated anatomical labeling atlas was established using partial correlation coefficients, and the whole-brain functional connectome was constructed by applying a threshold to the resultant 90 * 90 partial correlation matrix. Graph theory analysis was used to examine the topological properties of the functional connectome. A DL algorithm then used this measure to classify pediatric PTSD vs HC. Results
Graphic topological measures using DL provide a potentially clinically useful classifier for differentiating pediatric PTSD and HC (overall accuracy 71.2%). Frontoparietal areas (central executive network), cingulate cortex, and amygdala contributed the most to the DL model’s performance. Conclusions
Graphic topological measures based on fMRI data could contribute to imaging models of clinical utility in distinguishing pediatric PTSD from HC. DL model may be a useful tool in the identification of brain mechanisms PTSD participants. Supplementary Information
The online version contains supplementary material available at 10.1186/s12888-021-03503-9.
SUBMITTER: Yang J
PROVIDER: S-EPMC8555083 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Trauma-focused psychotherapy is the first-line treatment for posttraumatic stress disorder (PTSD) but 30-50% of patients do not benefit sufficiently. We investigated whether structural and resting-state functional magnetic resonance imaging (MRI/rs-fMRI) data could distinguish between treatment responders and non-responders on the group and individual level. Forty-four male veterans with PTSD underwent baseline scanning followed by trauma-focused psychotherapy. Voxel-wise gray matter volumes were extracted from the structural MRI data and resting-state networks (RSNs) were calculated from rs-fMRI data using independent component analysis. Data were used to detect differences between responders and non-responders on the group level using permutation testing, and the single-subject level using Gaussian process classification with cross-validation. A RSN centered on the bilateral superior frontal gyrus differed between responders and non-responder groups (PFWE < 0.05) while a RSN centered on the pre-supplementary motor area distinguished between responders and non-responders on an individual-level with 81.4% accuracy (P < 0.001, 84.8% sensitivity, 78% specificity and AUC of 0.93). No significant single-subject classification or group differences were observed for gray matter volume. This proof-of-concept study demonstrates the feasibility of using rs-fMRI to develop neuroimaging biomarkers for treatment response, which could enable personalized treatment of patients with PTSD.
| S-EPMC6889413 | biostudies-literature
Project description:ObjectiveResting-state functional magnetic resonance imaging (rs-fMRI) studies of adult posttraumatic stress disorder (PTSD) have identified default-mode network (DMN) abnormalities, including reduced within-network connectivity and reduced anticorrelation between the DMN and task-positive network (TPN). However, no prior studies have specifically examined DMN connectivity in pediatric PTSD, which may differ due to neurodevelopmental factors.MethodA total of 29 youth with PTSD and 30 nontraumatized healthy youth of comparable age and sex completed rs-fMRI. DMN properties were examined using posterior cingulate cortex (PCC) seed-based connectivity and independent component analysis (ICA).ResultsContrary to findings in adult studies, youth with PTSD displayed increased connectivity within the DMN, including increased PCC-inferior parietal gyrus connectivity, and age-related increases in PCC-ventromedial prefrontal cortex connectivity. Strikingly, youth with PTSD also displayed greater anticorrelation between the PCC and multiple nodes within salience and attentional control networks of the TPN. ICA revealed greater anticorrelation between the entire DMN and TPN networks in youth with PTSD. Furthermore, DMN and TPN connectivity strength were positively and negatively associated, respectively, with re-experiencing symptoms of PTSD.ConclusionPediatric PTSD is characterized by heightened within-DMN connectivity, which may contribute to re-experiencing symptoms of PTSD and is consistent with the role of the DMN in autobiographical memory. At the same time, greater anticorrelation between the DMN and attentional control networks may represent compensatory mechanisms aimed at suppressing trauma-related thought, a notion supported by the inverse relationship between TPN strength and re-experiencing. These findings provide new insights into large-scale network abnormalities underlying pediatric PTSD, which could serve as biomarkers of illness and treatment response.
| S-EPMC4808564 | biostudies-literature
Project description:The implementation of intelligent software to identify and classify objects and individuals in visual fields is a technology of growing importance to operatives in many fields, including wildlife conservation and management. To non-experts, the methods can be abstruse and the results mystifying. Here, in the context of applying cutting edge methods to classify wildlife species from camera-trap data, we shed light on the methods themselves and types of features these methods extract to make efficient identifications and reliable classifications. The current state of the art is to employ convolutional neural networks (CNN) encoded within deep-learning algorithms. We outline these methods and present results obtained in training a CNN to classify 20 African wildlife species with an overall accuracy of 87.5% from a dataset containing 111,467 images. We demonstrate the application of a gradient-weighted class-activation-mapping (Grad-CAM) procedure to extract the most salient pixels in the final convolution layer. We show that these pixels highlight features in particular images that in some cases are similar to those used to train humans to identify these species. Further, we used mutual information methods to identify the neurons in the final convolution layer that consistently respond most strongly across a set of images of one particular species. We then interpret the features in the image where the strongest responses occur, and present dataset biases that were revealed by these extracted features. We also used hierarchical clustering of feature vectors (i.e., the state of the final fully-connected layer in the CNN) associated with each image to produce a visual similarity dendrogram of identified species. Finally, we evaluated the relative unfamiliarity of images that were not part of the training set when these images were one of the 20 species "known" to our CNN in contrast to images of the species that were "unknown" to our CNN.
| S-EPMC6544615 | biostudies-literature
Project description:The necessary requirement of a traumatic event preceding the development of Posttraumatic Stress Disorder, theoretically allows for administering preventive and early interventions in the early aftermath of such events. Machine learning models including biomedical data to forecast PTSD outcome after trauma are highly promising for detection of individuals most in need of such interventions. In the current study, machine learning was applied on biomedical data collected within 48 h post-trauma to forecast individual risk for long-term PTSD, using a multinominal approach including the full spectrum of common PTSD symptom courses within one prognostic model for the first time. N = 417 patients (37.2% females; mean age 46.09 ± 15.88) admitted with (suspected) serious injury to two urban Academic Level-1 Trauma Centers were included. Routinely collected biomedical information (endocrine measures, vital signs, pharmacotherapy, demographics, injury and trauma characteristics) upon ED admission and subsequent 48 h was used. Cross-validated multi-nominal classification of longitudinal self-reported symptom severity (IES-R) over 12 months and bimodal classification of clinician-rated PTSD diagnosis (CAPS-IV) at 12 months post-trauma was performed using extreme Gradient Boosting and evaluated on hold-out sets. SHapley Additive exPlanations (SHAP) values were used to explain the derived models in human-interpretable form. Good prediction of longitudinal PTSD symptom trajectories (multiclass AUC = 0.89) and clinician-rated PTSD at 12 months (AUC = 0.89) was achieved. Most relevant prognostic variables to forecast both multinominal and dichotomous PTSD outcomes included acute endocrine and psychophysiological measures and hospital-prescribed pharmacotherapy. Thus, individual risk for long-term PTSD was accurately forecasted from biomedical information routinely collected within 48 h post-trauma. These results facilitate future targeted preventive interventions by enabling future early risk detection and provide further insights into the complex etiology of PTSD.
| S-EPMC7843920 | biostudies-literature
Project description:Functional abnormalities in fear circuitry are likely to underlie the pathophysiology of pediatric posttraumatic stress disorder (PTSD), but the few studies to date have yielded conflicting findings. Furthermore, network level functional connectivity and age-related disruptions in fear circuitry have not been thoroughly explored. In a cross-sectional design, 24 healthy and 24 medication-free youth with severe PTSD completed an event-related emotion-processing task during functional MRI. Youth viewed threat and neutral images, half of which were paired with a neutral male face. Group- and age-related differences in brain activation were examined in the medial prefrontal cortex (mPFC), amygdala, and hippocampus. Amygdala functional connectivity was examined using a seed-based approach. PTSD youth showed hyperactivation of the dorsal anterior cingulate cortex (dACC) to threat images. In the dorsomedial PFC (dmPFC), age positively predicted activation in healthy youth but negatively predicted activation in PTSD youth. In the amygdala functional connectivity analysis, PTSD youth showed decreased amygdala-mPFC connectivity to threat images. Furthermore, age positively predicted amygdala-vmPFC connectivity in healthy youth, but negatively predicted connectivity in PTSD youth. Finally, dmPFC activation and amygdala-mPFC connectivity were inversely related to PTSD severity. Pediatric PTSD involves abnormal functional activation and connectivity in fear circuitry. Specifically, dACC hyperactivation is consistent with abnormal promotion of fear responses, whereas reduced amygdala-mPFC connectivity suggests impaired regulation of amygdala responses to threat. Importantly, age-dependent decreases in dmPFC activation and amygdala-vmPFC connectivity may indicate abnormal developmental processes in key emotion pathways in pediatric PTSD.
| S-EPMC4707828 | biostudies-literature
Project description:Riboswitches are cis-regulatory genetic elements that use an aptamer to control gene expression. Specificity to cognate ligand and diversity of such ligands have expanded the functional repertoire of riboswitches to mediate mounting apt responses to sudden metabolic demands and signal changes in environmental conditions. Given their critical role in microbial life, riboswitch characterisation remains a challenging computational problem. Here we have addressed the issue with advanced deep learning frameworks, namely convolutional neural networks (CNN), and bidirectional recurrent neural networks (RNN) with Long Short-Term Memory (LSTM). Using a comprehensive dataset of 32 ligand classes and a stratified train-validate-test approach, we demonstrated the accurate performance of both the deep learning models (CNN and RNN) relative to conventional hyperparameter-optimized machine learning classifiers on all key performance metrics, including the ROC curve analysis. In particular, the bidirectional LSTM RNN emerged as the best-performing learning method for identifying the ligand-specificity of riboswitches with an accuracy >0.99 and macro-averaged F-score of 0.96. An additional attraction is that the deep learning models do not require prior feature engineering. A dynamic update functionality is built into the models to factor for the constant discovery of new riboswitches, and extend the predictive modeling to new classes. Our work would enable the design of genetic circuits with custom-tuned riboswitch aptamers that would effect precise translational control in synthetic biology. The associated software is available as an open-source Python package and standalone resource for use in genome annotation, synthetic biology, and biotechnology workflows.
| S-EPMC7371854 | biostudies-literature
Project description:Although medications and psychotherapy are often effective for the treatment of posttraumatic stress disorder (PTSD), 20-30% of patients do not respond to these conventional therapies. In psychiatry, DBS has been either approved or is currently under investigation for different disorders. At present, whether DBS may be used to treat PTSD remains unknown. Preclinical research may provide the scientific rationale for helping conceive and further improve such trials. Some of the animal models commonly used to date are more suitable for investigating mechanisms of anxiety and fear than the long-lasting behavior that characterized PTSD. That said, mechanisms and neurocircuits involved in paradigms such as fear conditioning and extinction share several common features with those of PTSD. In this article, we review preclinical studies in which electrical stimulation has been delivered to animal models of PTSD-like behavior. In those studies, commonly targeted regions were the basolateral amygdala, ventral striatum, hippocampus, and prefrontal cortex. Overall, stimulation delivered at high frequencies to most of these targets improved fear extinction and anxiety-like behavior. Though further research is certainly needed, promising findings from DBS studies in animals are encouraging and suggest a positive future perspective for the field.
| S-EPMC5061888 | biostudies-other
Project description:Disproportionate reactions to unexpected stimuli in the environment are a cardinal symptom of posttraumatic stress disorder (PTSD). Here, we test whether these heightened responses are associated with disruptions in distinct components of reinforcement learning. Specifically, using functional neuroimaging, a loss-learning task, and a computational model-based approach, we assessed the mechanistic hypothesis that overreactions to stimuli in PTSD arise from anomalous gating of attention during learning (i.e., associability). Behavioral choices of combat-deployed veterans with and without PTSD were fit to a reinforcement learning model, generating trial-by-trial prediction errors (signaling unexpected outcomes) and associability values (signaling attention allocation to the unexpected outcomes). Neural substrates of associability value and behavioral parameter estimates of associability updating, but not prediction error, increased with PTSD during loss learning. Moreover, the interaction of PTSD severity with neural markers of associability value predicted behavioral choices. These results indicate that increased attention-based learning may underlie aspects of PTSD and suggest potential neuromechanistic treatment targets.
| S-EPMC5760201 | biostudies-literature
Project description:Posttraumatic stress disorder (PTSD) may develop when mechanisms for making accurate distinctions about threat relevance have gone awry. Generalization across conceptually related objects has been hypothesized based on clinical observation in PTSD, but the neural mechanisms remain unexplored. Recent trauma-exposed military veterans (n = 46) were grouped into PTSD (n = 23) and non-PTSD (n = 23). Participants learned to generalize fear across conceptual categories (animals or tools) of semantically related items that were partially reinforced by shock during functional magnetic resonance imaging. Conditioned fear learning was quantified by shock expectancy and skin conductance response (SCR). Relative to veteran controls, PTSD subjects exhibited a stronger neural response associated with fear generalization to the reinforced object category in the striatum, anterior cingulate cortex, amygdala, occipitotemporal cortex, and insula (Z > 2.3; p < 0.05; whole-brain corrected). Based on SCR, both groups generalized the shock contingency to the reinforced conceptual category, but learning was not significantly different between groups. We found that PTSD was associated with an enhanced neural response in fronto-limbic, midline, and occipitotemporal regions to a learned representation of threat that is based on previously established conceptual knowledge of the relationship between basic-level exemplars within a semantic category. Behaviorally, veterans with PTSD were somewhat slower to differentiate threat and safety categories as compared with trauma-exposed veteran controls owing in part to an initial overgeneralized behavioral response to the safe category. These results have implications for understanding how fear spreads across semantically related concepts in PTSD.
| S-EPMC7297719 | biostudies-literature
Project description:BACKGROUND:Psychophysiological alterations such as elevated baseline levels and hyperresponsivity in cardiac, electrodermal, and facial muscle activity have been observed in adults with posttraumatic stress disorder (PTSD). There are only few, inconclusive studies investigating psychophysiological responses in children and adolescents with PTSD. OBJECTIVE:This cross-sectional study sought to examine if autonomic variables, facial electromyography (EMG), and self-reported anxiety at baseline, while listening to neutral and idiosyncratic trauma scripts, differ between minors with a trauma history and PTSD, and a traumatized control (TC) group without PTSD. A better understanding of psychophysiological reactions in trauma-exposed children and adolescents could improve differential assessment and treatment decisions. METHOD:PTSD was assessed using the Clinician Administered PTSD Scale for Children and Adolescents in 6- to 17-year-old trauma-exposed children, resulting in a group with PTSD according to DSM-IV (n=16) and a TC group without PTSD (n=18). Facial EMG, (para-)sympathetic measures (heart rate, electrodermal activity, respiratory sinus arrhythmia), and self-reported anxiety were measured during 5-min baseline, 3-min neutral script, and 3-min idiosyncratic trauma script. Baseline, reactivity (trauma minus baseline), and script contrast (trauma minus neutral) were analyzed by multivariate analyses of variance. RESULTS:Children and adolescents with PTSD reported more anxiety compared to TC for baseline, reactivity, and script contrast (ps<0.021, ds>0.59), and showed elevated corrugator supercilii muscle activity for script contrast (p<0.05, d=0.79). No group differences emerged for sympathetic or parasympathetic measures. CONCLUSIONS:Children and adolescents with PTSD experienced elevated anxiety at baseline and elevated anxiety and facial corrugator muscle response to an idiosyncratic trauma narrative. Autonomic hyperreactivity, typical for adult PTSD samples, did not figure prominently.
| S-EPMC4320135 | biostudies-literature