Impact of veteran-led peer mentorship on posttraumatic stress disorder.

Peer mentorship shows promise as a strategy to support veteran mental health. A community-academic partnership involving a veteran-led nonprofit organization and institutions of higher education evaluated a collaboratively developed peer mentor intervention. We assessed posttraumatic stress disorder (PTSD), postdeployment experiences, social functioning, and psychological strengths at baseline, midpoint, and 12-week discharge using the PTSD Checklist for DSM-5 (PCL-5), Deployment Risk and Resilience Inventory-2, Social Adaptation Self-evaluation Scale, and Values in Action Survey. Brief weekly check-in surveys reinforced mentor contact and assessed retention. The sample included 307 veterans who were served by 17 veteran peer mentors. Mixed-effects linear models found a modest effect for PTSD symptom change, with a mean PCL-5 score reduction of 4.04 points, 95% CI [-6.44, -1.64], d = 0.44. More symptomatic veterans showed a larger effect, with average reductions of 9.03 points, 95% CI [-12.11, -5.95], d = 0.77. There were no significant findings for other outcome variables. Compared to younger veterans, those aged 32-57 years were less likely to drop out by 6 weeks, aORs = 0.32-0.26. Week-by-week hazard of drop-out was lower with mentors ≥ 35 years old, aHR = 0.62, 95% CI [0.37, 1.05]. Unadjusted survival differed by mentor military branch, p = .028, but the small mentor sample reduced interpretability. Like many community research efforts, this study lacked a control group, limiting the inferences that can be drawn. Continued study of veteran peer mentorship is important as this modality is often viewed as more tolerable than therapy.

Data augmentation strategies for EEG-based motor imagery decoding.

The wide use of motor imagery as a paradigm for brain-computer interfacing (BCI) points to its characteristic ability to generate discriminatory signals for communication and control. In recent times, deep learning techniques have increasingly been explored, in motor imagery decoding. While deep learning techniques are promising, a major challenge limiting their wide adoption is the amount of data available for decoding. To combat this challenge, data augmentation can be performed, to enhance decoding performance. In this study, we performed data augmentation by synthesizing motor imagery (MI) electroencephalography (EEG) trials, following six approaches. Data generated using these methods were evaluated based on four criteria, namely - the accuracy of prediction, the Frechet Inception distance (FID), the t-distributed Stochastic Neighbour Embedding (t-SNE) plots and topographic head plots. We show, based on these, that the synthesized data exhibit similar characteristics with real data, gaining up to 3% and 12% increases in mean accuracies across two public datasets. Finally, we believe these approaches should be utilized in applying deep learning techniques, as they not only have the potential to improve prediction performances, but also to save time spent on subject data collection.