Abstract:Feature selection can disclose biomarkers of mental disorders that have unclear biological mechanisms. Although neighborhood rough set (NRS) has been applied to discover important sparse features, it has hardly ever been utilized in neuroimaging-based biomarker identification, probably due to the inadequate feature evaluation metric and incomplete information provided under a single-granularity. Here, we propose a new NRS-based feature selection method and successfully identify brain functional connectivity biomarkers of schizophrenia (SZ) using functional magnetic resonance imaging (fMRI) data. Specifically, we develop a new weighted metric based on NRS combined with information entropy to evaluate the capacity of features in distinguishing different groups. Inspired by multi-granularity information maximization theory, we further take advantage of the complementary information from different neighborhood sizes via a multi-granularity fusion to obtain the most discriminative and stable features. For validation, we compare our method with six popular feature selection methods using three public omics datasets as well as resting-state fMRI data of 393 SZ patients and 429 healthy controls. Results show that our method obtained higher classification accuracies on both omics data (100.0%, 88.6%, and 72.2% for three omics datasets, respectively) and fMRI data (93.9% for main dataset, and 76.3% and 83.8% for two independent datasets, respectively). Moreover, our findings reveal biologically meaningful substrates of SZ, notably involving the connectivity between the thalamus and superior temporal gyrus as well as between the postcentral gyrus and calcarine gyrus. Taken together, we propose a new NRS-based feature selection method that shows the potential of exploring effective and sparse neuroimaging-based biomarkers of mental disorders.

Hippocampal shape classification using redundancy constrained feature selection.

Shape Representation and Recognition in High Dimensional Feature Space

Balancing the Stability and Predictive Performance for Multivariate Voxel Selection in fMRI Study.

A Novel Approach for Stable Selection of Informative Redundant Features from High Dimensional fMRI Data

Feature Selection with Redundancy-Constrained Class Separability

A Unified Feature Representation and Learning Framework for 3D Shape

A Dimensional Reduction Guiding Deep Learning Architecture for 3D Shape Retrieval

Clustering-based feature subset selection with analysis on the redundancy–complementarity dimension

A Constrained Feature Selection Approach Based on Feature Clustering and Hypothesis Margin Maximization

A Feature Selection Method Based on Feature Grouping and Genetic Algorithm

Local-structure-preservation and redundancy-removal-based feature selection method and its application to the identification of biomarkers for schizophrenia

A Novel Neighborhood Rough Set-Based Feature Selection Method and Its Application to Biomarker Identification of Schizophrenia

Feature Selection based on Machine Learning in MRIs for Hippocampal Segmentation

Quantifying Hippocampal Shape Asymmetry in Alzheimer's Disease Using Optimal Shape Correspondences

Feature Selection Based on Orthogonal Constraints and Polygon Area

Multidimensional classification of hippocampal shape features discriminates Alzheimer's disease and mild cognitive impairment from normal aging

Supervised Feature Selection With Orthogonal Regression and Mountain Range Area

Supervised redundant feature detection for tumor classification

Automatically Redundant Features Removal for Unsupervised Feature Selection via Sparse Feature Graph

Adaptive Feature Selection With Augmented Attributes

Compartmental Sparse Feature Selection Method for Alzheimer's Disease Identification.