Traumatic brain injury (TBI) disrupts brain communication and increases risk for post-traumatic stress disorder (PTSD). However, mechanisms by which TBI-related disruption of brain communication confers PTSD risk have not been successfully elucidated in humans. This may be in part because functional MRI (fMRI), the most common technique for measuring functional brain communication, is unreliable for characterizing individual patients. However, this unreliability can be overcome with sufficient within-individual data. Here, we examined whether relationships could be observed among TBI, structural and functional brain connectivity, and PTSD severity by collecting ∼3.5 hours of resting-state fMRI and diffusion tensor imaging (DTI) data in each of 26 United States military veterans. We observed that a TBI history was associated with decreased whole-brain resting-state functional connectivity (RSFC), while the number of lifetime TBIs was associated with reduced whole-brain fractional anisotropy (FA). Both RSFC and FA explained independent variance in PTSD severity, with RSFC mediating the TBI-PTSD relationship. Finally, we showed that large amounts of per-individual data produced highly reliable RSFC measures, and that relationships among TBI, RSFC/FA, and PTSD could not be observed with typical data quantities. These results demonstrate links among TBI, brain connectivity, and PTSD severity, and illustrate the need for precise characterization of individual patients using high-data fMRI scanning.
Background: Current clinical assessments of Posttraumatic stress disorder (PTSD) rely solely on subjective symptoms and experiences reported by the patient, rather than objective biomarkers of the illness. Recent advances in data-driven computational approaches have been helpful in devising tools to objectively diagnose psychiatric disorders. Here we aimed to classify individuals with PTSD versus controls using heterogeneous brain datasets from the ENIGMA-PGC PTSD Working group. Methods: We analyzed brain MRI data from 3,527 structural-MRI; 2,502 resting state-fMRI; and 1,953 diffusion-MRI. First, we identified the brain features that best distinguish individuals with PTSD from controls (TEHC and HC) using traditional machine learning methods. Second, we assessed the utility of the denoising variational autoencoder (DVAE) and evaluated its classification performance. Third, we assessed the generalizability and reproducibility of both models using leave-one-site-out cross-validation procedure for each modality. Results: We found lower performance in classifying PTSD vs. controls with data from over 20 sites (60% test AUC for s-MRI, 59% for rs-fMRI and 56% for d-MRI), as compared to other studies run on single-site data. The performance increased when classifying PTSD from HC without trauma history across all three modalities (75% AUC). The classification performance remained intact when applying the DVAE framework, which reduced the number of features. Finally, we found that the DVAE framework achieved better generalization to unseen datasets compared with the traditional machine learning frameworks, albeit performance was slightly above chance. Conclusion: Our findings highlight the promise offered by machine learning methods for the diagnosis of patients with PTSD. The utility of brain biomarkers across three MRI modalities and the contribution of DVAE models for improving generalizability offers new insights into neural mechanisms involved in PTSD.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.