An ensemble multimodal approach for predicting first episode psychosis using structural MRI and cognitive assessments
Zhang, S.
Show abstract
Classification between first episode psychosis (FEP) patients and healthy controls is of particular interest to the study of schizophrenia. However, predicting psychosis with cognitive assessments alone is prone to human errors and often lacks biological evidence to back up the findings. In this work, we combined a multimodal dataset of structural MRI and cognitive data to disentangle the detection of first-episode psychosis with a machine learning approach. For this purpose, we proposed a robust detection pipeline that explores the variables in high-order feature space. We applied the pipeline to Human Connectome Project for Early Psychosis (HCP-EP) dataset with 108 participants in EP and 47 controls. The pipeline demonstrated strong performance with 74.67% balanced accuracy on this task. Further feature analysis shows that the model is capable of identifying verified causative biological factors for the occurrence of psychosis based on volumetric MRI measurements, which suggests the potential of data-driven approaches for the search for neuroimaging biomarkers in future studies.
Matching journals
The top 3 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Bridging Structural MRI with Cognitive Function for Individual Level Classification of Early Psychosis via Deep Learning 97%
- Explainable Fuzzy Clustering Framework Reveals Divergent Default Mode Network Connectivity Dynamics in Schizophrenia 94%
- Understanding Psychiatric Illness Through Natural Language Processing (UNDERPIN): Rationale, Design, and Methodology 93%
Similar papers in this journal
- Who Does What to Whom? Graph Representations of Action-Predication in Speech Relate to Psychopathological Dimensions of Psychosis 92%
- Multimodal fusion of brain signals for robust prediction of psychosis transition 91%
- Temporal characteristics of hemodynamic responses during active and passive hand movements in schizophrenia spectrum disorder 91%
Similar papers in this journal
- A Multimodal Vision Transformer for InterpretableFusion of Functional and Structural NeuroimagingData 97%
- Time-varying Spatial Propagation of Brain Networks in fMRI data 94%
- A method for estimating dynamic functional network connectivity gradients (dFNG) from ICA captures smooth inter-network modulation. 94%
Similar papers in this journal
- Providing context: Extracting non-linear and dynamic temporal motifs from brain activity 95%
- A Confounder Controlled Machine Learning Approach: Group Analysis and Classification of Schizophrenia and Alzheimer's Disease using Resting-State Functional Network Connectivity 93%
- Multiclass Classification of Autism Spectrum Disorder, Attention Deficit Hyperactivity Disorder, and Typically Developed Individuals Using fMRI Functional Connectivity Analysis 93%
Similar papers in this journal
- Deep learning approach for automatic assessment of schizophrenia and bipolar disorder in patients using R-R intervals 93%
- Controlling astrocyte-mediated synaptic pruning signals for schizophrenia drug repurposing with Deep Graph Networks 91%
- A novel classification framework for genome-wide association study of whole brain MRI images using deep learning 90%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.