Using genetic, cognitive and multi-modal neuroimaging data to identify ultra-high-risk and first-episode psychosis at the individual level

Abstract

BackgroundGroup-level results suggest that relative to healthy controls (HCs), ultra-high-risk (UHR) and first-episode psychosis (FEP) subjects show alterations in neuroanatomy, neurofunction and cognition that may be mediated genetically. It is unclear, however, whether these groups can be differentiated at single-subject level, for instance using the machine learning analysis support vector machine (SVM). Here, we used a multimodal approach to examine the ability of structural magnetic resonance imaging (sMRI), functional MRI (fMRI), diffusion tensor neuroimaging (DTI), genetic and cognitive data to differentiate between UHR, FEP and HC subjects at the single-subject level using SVM.MethodThree age- and gender-matched SVM paired comparison groups were created comprising 19, 19 and 15 subject pairs for FEP versus HC, UHR versus HC and FEP versus UHR, respectively. Genetic, sMRI, DTI, fMRI and cognitive data were obtained for each participant and the ability of each to discriminate subjects at the individual level in conjunction with SVM was tested.ResultsSuccessful classification accuracies (p < 0.05) comprised FEP versus HC (genotype, 67.86%; DTI, 65.79%; fMRI, 65.79% and 68.42%; cognitive data, 73.69%), UHR versus HC (sMRI, 68.42%; DTI, 65.79%), and FEP versus UHR (sMRI, 76.67%; fMRI, 73.33%; cognitive data, 66.67%).ConclusionsThe results suggest that FEP subjects are identifiable at the individual level using a range of biological and cognitive measures. Comparatively, only sMRI and DTI allowed discrimination of UHR from HC subjects. For the first time FEP and UHR subjects have been shown to be directly differentiable at the single-subject level using cognitive, sMRI and fMRI data. Preliminarily, the results support clinical development of SVM to help inform identification of FEP and UHR subjects, though future work is needed to provide enhanced levels of accuracy.

The use of dynamic susceptibility contrast (DSC) MRI to automatically classify patients with first episode psychosis

“…However, our study also suffers from some limitations. The sample size is relatively limited, although in line with other studies using SVM combined with MRI (Li et al, 2014;Pettersson-Yeo et al, 2013). Moreover, the DSC method is inherently invasive due to tracer injection thus limiting its use in routine clinical practice (although it allows the direct estimation of brain perfusion parameters).…”

Diffusion tensor imaging in first degree relatives of schizophrenia and bipolar disorder patients

“…We included one study that enrolled not only unaffected but also prodromal relatives (Hoptman et al, 2008). We excluded otherwise relevant studies if they didn't compare the groups in terms of DTI values (Bertisch et al, 2010), only focused on individuals at UHR or clinical high risk for psychosis (Carletti et al, 2012; Jacobson et al, 2010; Karlsgodt et al, 2009; Peters et al, 2008; Pettersson-Yeo et al, 2013; von Hohenberg et al, 2014), applied DTI to measure ADC in grey matter (Narr et al, 2009), were not written in English (Kang et al, 2012) or was a poster presented at a meeting and not a published manuscript (Contet et al, 2011). …”

Characterization of major depressive disorder using a multiparametric classification approach based on high resolution structural images

“…It has been successfully used to distinguish between trauma survivors with and without posttraumatic stress disorder; 43 predict responses to antidepressant medications; 27 and identify ultra-high risk and first-episode psychosis based on neuroimaging, cognitive and genetic data. 44 The PROBID software allows for a linear kernel matrix (which measures the similarity between all pairs of brain images) to be precomputed and supplied to the classifier. This approach affords a substantial increase in computational efficiency and permits whole brain classification without requiring explicit dimensionality reduction.…”