1
|
DeRamus TP, Wu L, Qi S, Iraji A, Silva R, Du Y, Pearlson G, Mayer A, Bustillo JR, Stromberg SF, Calhoun VD. Multimodal data fusion of cortical-subcortical morphology and functional network connectivity in psychotic spectrum disorder. Neuroimage Clin 2022; 35:103056. [PMID: 35709557 PMCID: PMC9207350 DOI: 10.1016/j.nicl.2022.103056] [Citation(s) in RCA: 4] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/10/2022] [Revised: 04/18/2022] [Accepted: 05/21/2022] [Indexed: 11/20/2022]
Abstract
Overlap has been noted disorders which fall on the psychotic spectrum. Univariate studies may miss joint brain features across diagnostic categories. mCCA with jICA is paired with features across the psychotic spectrum to produce joint components. One joint component displayed a significant relationship with cognitive scores. The replicate trends of cortical-subcortical irregularity in psychotic spectrum disorders.
Multiple authors have noted overlapping symptoms and alterations across clinical, anatomical, and functional brain features in schizophrenia (SZ), schizoaffective disorder (SZA), and bipolar disorder (BPI). However, regarding brain features, few studies have approached this line of inquiry using analytical techniques optimally designed to extract the shared features across anatomical and functional information in a simultaneous manner. Univariate studies of anatomical or functional alterations across these disorders can be limited and run the risk of omitting small but potentially crucial overlapping or joint neuroanatomical (e.g., structural images) and functional features (e.g., fMRI-based features) which may serve as informative clinical indicators of across multiple diagnostic categories. To address this limitation, we paired an unsupervised multimodal canonical correlation analysis (mCCA) together with joint independent component analysis (jICA) to identify linked spatial gray matter (GM), resting-state functional network connectivity (FNC), and white matter fractional anisotropy (FA) features across these diagnostic categories. We then calculated associations between the identified linked features and trans-diagnostic behavioral measures (MATRICs Consensus Cognitive Battery, MCCB). Component number 4 of the 13 identified displayed a statistically significant relationship with overall MCCB scores across GM, resting-state FNC, and FA. These linked modalities of component 4 consisted primarily of positive correlations within subcortical structures including the caudate and putamen in the GM maps with overall MCCB, sparse negative correlations within subcortical and cortical connection tracts (e.g., corticospinal tract, superior longitudinal fasciculus) in the FA maps with overall MCCB, and negative relationships with MCCB values and loading parameters with FNC matrices displaying increased FNC in subcortical-cortical regions with auditory, somatomotor, and visual regions.
Collapse
Affiliation(s)
- T P DeRamus
- Tri-institutional Center for Translational Research in Neuroimaging and Data Science (TReNDS) - Georgia State University, Georgia Institute of Technology, and Emory University, Atlanta, GA, USA.
| | - L Wu
- Tri-institutional Center for Translational Research in Neuroimaging and Data Science (TReNDS) - Georgia State University, Georgia Institute of Technology, and Emory University, Atlanta, GA, USA
| | - S Qi
- College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing, China
| | - A Iraji
- Tri-institutional Center for Translational Research in Neuroimaging and Data Science (TReNDS) - Georgia State University, Georgia Institute of Technology, and Emory University, Atlanta, GA, USA
| | - R Silva
- Tri-institutional Center for Translational Research in Neuroimaging and Data Science (TReNDS) - Georgia State University, Georgia Institute of Technology, and Emory University, Atlanta, GA, USA
| | - Y Du
- Tri-institutional Center for Translational Research in Neuroimaging and Data Science (TReNDS) - Georgia State University, Georgia Institute of Technology, and Emory University, Atlanta, GA, USA; School of Computer and Information Technology, Shanxi University, Taiyuan, China
| | - G Pearlson
- Olin Neuropsychiatry Research Center, Institute of Living at Hartford Hospital, Hartford, CT, USA; Department of Psychiatry, Yale University School of Medicine, New Haven, CT, USA; Department of Neuroscience, Yale University School of Medicine, New Haven, CT, USA
| | - A Mayer
- The Mind Research Network, Lovelace Biomedical and Environmental Research Institute, Albuquerque, USA
| | - J R Bustillo
- Department of Psychiatry, University of New Mexico School of Medicine, Albuquerque, NM, USA
| | - S F Stromberg
- Psychiatry and Behavioral Health Clinical Program, Presbyterian Healthcare System, Albuquerque, NM, USA
| | - V D Calhoun
- Tri-institutional Center for Translational Research in Neuroimaging and Data Science (TReNDS) - Georgia State University, Georgia Institute of Technology, and Emory University, Atlanta, GA, USA; The Mind Research Network, Lovelace Biomedical and Environmental Research Institute, Albuquerque, USA; Department of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, USA; Department of Computer Science, Georgia State University, Atlanta, USA; Department of Psychology, Georgia State University, Atlanta, USA
| |
Collapse
|
2
|
Clark DG, Kapur P, Geldmacher DS, Brockington JC, Harrell L, DeRamus TP, Blanton PD, Lokken K, Nicholas AP, Marson DC. Latent information in fluency lists predicts functional decline in persons at risk for Alzheimer disease. Cortex 2014; 55:202-18. [PMID: 24556551 PMCID: PMC4039569 DOI: 10.1016/j.cortex.2013.12.013] [Citation(s) in RCA: 34] [Impact Index Per Article: 3.4] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/12/2013] [Revised: 11/05/2013] [Accepted: 12/25/2013] [Indexed: 01/24/2023]
Abstract
OBJECTIVE We constructed random forest classifiers employing either the traditional method of scoring semantic fluency word lists or new methods. These classifiers were then compared in terms of their ability to diagnose Alzheimer disease (AD) or to prognosticate among individuals along the continuum from cognitively normal (CN) through mild cognitive impairment (MCI) to AD. METHOD Semantic fluency lists from 44 cognitively normal elderly individuals, 80 MCI patients, and 41 AD patients were transcribed into electronic text files and scored by four methods: traditional raw scores, clustering and switching scores, "generalized" versions of clustering and switching, and a method based on independent components analysis (ICA). Random forest classifiers based on raw scores were compared to "augmented" classifiers that incorporated newer scoring methods. Outcome variables included AD diagnosis at baseline, MCI conversion, increase in Clinical Dementia Rating-Sum of Boxes (CDR-SOB) score, or decrease in Financial Capacity Instrument (FCI) score. Receiver operating characteristic (ROC) curves were constructed for each classifier and the area under the curve (AUC) was calculated. We compared AUC between raw and augmented classifiers using Delong's test and assessed validity and reliability of the augmented classifier. RESULTS Augmented classifiers outperformed classifiers based on raw scores for the outcome measures AD diagnosis (AUC .97 vs. .95), MCI conversion (AUC .91 vs. .77), CDR-SOB increase (AUC .90 vs. .79), and FCI decrease (AUC .89 vs. .72). Measures of validity and stability over time support the use of the method. CONCLUSION Latent information in semantic fluency word lists is useful for predicting cognitive and functional decline among elderly individuals at increased risk for developing AD. Modern machine learning methods may incorporate latent information to enhance the diagnostic value of semantic fluency raw scores. These methods could yield information valuable for patient care and clinical trial design with a relatively small investment of time and money.
Collapse
Affiliation(s)
- D G Clark
- Birmingham VA Medical Center, USA; Department of Neurology, University of Alabama at Birmingham, USA.
| | - P Kapur
- Department of Biomedical Sciences, Georgia Health Science University, USA
| | - D S Geldmacher
- Department of Neurology, University of Alabama at Birmingham, USA
| | - J C Brockington
- Department of Neurology, University of Alabama at Birmingham, USA
| | - L Harrell
- Department of Neurology, University of Alabama at Birmingham, USA
| | - T P DeRamus
- Department of Psychology and Behavioral Neuroscience, University of Alabama at Birmingham, USA
| | | | - K Lokken
- Birmingham VA Medical Center, USA
| | - A P Nicholas
- Birmingham VA Medical Center, USA; Department of Neurology, University of Alabama at Birmingham, USA
| | - D C Marson
- Department of Neurology, University of Alabama at Birmingham, USA
| |
Collapse
|