Neuroanatomical Classification in a Population-Based Sample of Psychotic Major Depression and Bipolar I Disorder with 1 Year of Diagnostic Stability

Abstract

The presence of psychotic features in the course of a depressive disorder is known to increase the risk for bipolarity, but the early identification of such cases remains challenging in clinical practice. In the present study, we evaluated the diagnostic performance of a neuroanatomical pattern classification method in the discrimination between psychotic major depressive disorder (MDD), bipolar I disorder (BD-I), and healthy controls (HC) using a homogenous sample of patients at an early course of their illness. Twenty-three cases of first-episode psychotic mania (BD-I) and 19 individuals with a first episode of psychotic MDD whose diagnosis remained stable during 1 year of followup underwent 1.5 T MRI at baseline. A previously validated multivariate classifier based on support vector machine (SVM) was employed and measures of diagnostic performance were obtained for the discrimination between each diagnostic group and subsamples of age- and gender-matched controls recruited in the same neighborhood of the patients. Based on T1-weighted images only, the SVM-classifier afforded poor discrimination in all 3 pairwise comparisons: BD-I versus HC; MDD versus HC; and BD-I versus MDD. Thus, at the population level and using structural MRI only, we failed to achieve good discrimination between BD-I, psychotic MDD, and HC in this proof of concept study.

Detecting Neuroimaging Biomarkers for Depression: A Meta-analysis of Multivariate Pattern Recognition Studies

“…In this context, a recent metaanalysis showed that patients with schizophrenia can be accurately differentiated from healthy volunteers in 80% of the cases using only neuroimaging-based diagnostic models (17). Moreover, these methods may facilitate the development of neuroimaging tools to distinguish among different psychiatric disorders (18)(19)(20)(21) or to predict clinical outcomes (22)(23)(24). Indeed, multiple proof-of-concept studies have successfully used multivariate statistical methods to guide the diagnosis of depression based on structural MRI (sMRI) data (19; 21; 25-27), resting-state functional MRI (rsfMRI) data (26; 28-34), and task-based functional MRI (fMRI) data (35)(36)(37)(38)(39).…”

Classification of MRI under the Presence of Disease Heterogeneity using Multi-Task Learning: Application to Bipolar Disorder

“…According to the DSM-IV criteria, each patient was assigned into BD I (22 subjects) or BD II (22 subjects) subgroups. Details on demographic characteristics, and image acquisition and preprocessing can be found in [16]. T1-weighted images were preprocessed according to a number of steps [16], including 1) AC-PC plane alignment; 2) Skull removal; 3) Tissue segmentation into gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF); and 4) High-dimensional image warping to a standard MNI space, resulting in the mass-preserved tissue density maps.…”

Consensus paper of the WFSBP Task Force on Biological Markers: Criteria for biomarkers and endophenotypes of schizophrenia part II: Cognition, neuroimaging and genetics

“…Classification accuracies of 67 and 80% have been reported for the differentiation between depressive and bipolar patients on the basis of MVPA of fMRI data (Mourao-Miranda et al 2012;Grotegerd et al 2014). One study reported only relatively low accuracies for MVPA of sMRI data (Serpa et al 2014), but a more recent study demonstrated 79% classification accuracy for the differentiation between patients with unipolar and bipolar depression , and another study showed 88% accuracy for the differentiation between bipolar patients and patients with schizophrenia (Schnack et al 2014). One DTI study reported accuracies of 72-88% when MVPA was used to differentiate healthy women from female schizophrenia patients (Ota et al 2012).…”