A Review of Feature Reduction Techniques in Neuroimaging

• Published in: Neuroinformatics (Totowa, N.J.) Vol. 12; no. 2; pp. 229 - 244
• Main Authors: Mwangi, Benson, Tian, Tian Siva, Soares, Jair C.
• Format: Journal Article
• Language: English
• Published: Boston Springer US 01.04.2014; Springer; Springer Nature B.V
• Subjects: Applied sciences; Artificial Intelligence; Bioinformatics; Biological and medical sciences; Biomedical and Life Sciences; Biomedicine; Brain - physiology; Computational Biology/Bioinformatics; Computer Appl. in Life Sciences; Computer science; control theory; systems; Data processing. List processing. Character string processing; Exact sciences and technology; Humans; Investigative techniques, diagnostic techniques (general aspects); Learning and adaptive systems; Medical sciences; Memory organisation. Data processing; Nervous system; Neuroimaging - methods; Neuroimaging - statistics & numerical data; Neurology; Neurosciences; Pattern recognition. Digital image processing. Computational geometry; Radiodiagnosis. Nmr imagery. Nmr spectrometry; Review; Software; Statistics as Topic; Dimensionality reduction; Predictive Modeling; Neuroimaging; Feature selection; Machine learning; Multivariate; Feature reduction; Central nervous system; Modeling; Encephalon; Multidimensional analysis; Voxel; Selection criterion; Age; Medical application; Computer vision; Data analysis; Redundancy; Inference; Data reduction; Dimension reduction; Discrimination; Medical imagery; Tridimensional image; Artificial intelligence
• ISSN: 1539-2791; 1559-0089; 1559-0089
• DOI: 10.1007/s12021-013-9204-3

Machine learning techniques are increasingly being used in making relevant predictions and inferences on individual subjects neuroimaging scan data. Previous studies have mostly focused on categorical discrimination of patients and matched healthy controls and more recently, on prediction of individual continuous variables such as clinical scores or age. However, these studies are greatly hampered by the large number of predictor variables (voxels) and low observations (subjects) also known as the curse-of-dimensionality or small-n-large-p problem. As a result, feature reduction techniques such as feature subset selection and dimensionality reduction are used to remove redundant predictor variables and experimental noise, a process which mitigates the curse-of-dimensionality and small-n-large-p effects. Feature reduction is an essential step before training a machine learning model to avoid overfitting and therefore improving model prediction accuracy and generalization ability. In this review, we discuss feature reduction techniques used with machine learning in neuroimaging studies.