Logo image
Back
Neuroimaging-based classification of PTSD using data-driven computational approaches: A multisite big data study from the ENIGMA-PGC PTSD consortium
Journal article   Open access   Peer reviewed

Neuroimaging-based classification of PTSD using data-driven computational approaches: A multisite big data study from the ENIGMA-PGC PTSD consortium

Xi Zhu, Yoojean Kim, Orren Ravid, Xiaofu He, Benjamin Suarez-Jimenez, Sigal Zilcha-Mano, Amit Lazarov, Seonjoo Lee, Chadi G. Abdallah, Michael Angstadt, …
NeuroImage (Orlando, Fla.), Vol.283, 120412
12/01/2023
DOI: 10.1016/j.neuroimage.2023.120412
PMCID: PMC10842116
PMID: 37858907
url
https://doi.org/10.1016/j.neuroimage.2023.120412View
Published (Version of record) Open Access

Abstract

•Classifying PTSD from trauma-unexposed healthy controls (HC) using three imaging modalities performed well (∼75 % AUC), but performance suffered markedly when classifying PTSD from trauma-exposed healthy controls (TEHC) using three imaging modalities (∼60 % AUC).•Using deep learning for feature reduction (denoising variational auto-encoder; DVAE) dramatically reduced the number of features with no concomitant performance degradation.•Utilizing denoising variational autoencoder (DVAE) models improves generalizability across heterogeneous multi-site data compared with the traditional machine learning frameworks. Recent advances in data-driven computational approaches have been helpful in devising tools to objectively diagnose psychiatric disorders. However, current machine learning studies limited to small homogeneous samples, different methodologies, and different imaging collection protocols, limit the ability to directly compare and generalize their results. Here we aimed to classify individuals with PTSD versus controls and assess the generalizability using a large heterogeneous brain datasets from the ENIGMA-PGC PTSD Working group. We analyzed brain MRI data from 3,477 structural-MRI; 2,495 resting state-fMRI; and 1,952 diffusion-MRI. First, we identified the brain features that best distinguish individuals with PTSD from controls 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. 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 in each modality (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. These results have the potential to provide a baseline classification performance for PTSD when using large scale neuroimaging datasets. Our findings show that the control group used can heavily affect classification performance. The DVAE framework provided better generalizability for the multi-site data. This may be more significant in clinical practice since the neuroimaging-based diagnostic DVAE classification models are much less site-specific, rendering them more generalizable.
 Machine Learning Posttraumatic Stress Disorder Classification Deep learning Multimodal MRI

Details

Metrics

10 Record Views
14 Times Cited - Web of Science
Logo image