White matter hyperintensity and stroke lesion segmentation and differentiation using convolutional neural networks.

White matter hyperintensities (WMH) are a feature of sporadic small vessel disease also frequently observed in magnetic resonance images (MRI) of healthy elderly subjects. The accurate assessment of WMH burden is of crucial importance for epidemiological studies to determine association between WMHs, cognitive and clinical data; their causes, and the effects of new treatments in randomized trials. The manual delineation of WMHs is a very tedious, costly and time consuming process, that needs to be carried out by an expert annotator (e.g. a trained image analyst or radiologist). The problem of WMH delineation is further complicated by the fact that other pathological features (i.e. stroke lesions) often also appear as hyperintense regions. Recently, several automated methods aiming to tackle the challenges of WMH segmentation have been proposed. Most of these methods have been specifically developed to segment WMH in MRI but cannot differentiate between WMHs and strokes. Other methods, capable of distinguishing between different pathologies in brain MRI, are not designed with simultaneous WMH and stroke segmentation in mind. Therefore, a task specific, reliable, fully automated method that can segment and differentiate between these two pathological manifestations on MRI has not yet been fully identified. In this work we propose to use a convolutional neural network (CNN) that is able to segment hyperintensities and differentiate between WMHs and stroke lesions. Specifically, we aim to distinguish between WMH pathologies from those caused by stroke lesions due to either cortical, large or small subcortical infarcts. The proposed fully convolutional CNN architecture, called uResNet, that comprised an analysis path, that gradually learns low and high level features, followed by a synthesis path, that gradually combines and up-samples the low and high level features into a class likelihood semantic segmentation. Quantitatively, the proposed CNN architecture is shown to outperform other well established and state-of-the-art algorithms in terms of overlap with manual expert annotations. Clinically, the extracted WMH volumes were found to correlate better with the Fazekas visual rating score than competing methods or the expert-annotated volumes. Additionally, a comparison of the associations found between clinical risk-factors and the WMH volumes generated by the proposed method, was found to be in line with the associations found with the expert-annotated volumes.

Instability of default mode network connectivity in major depression: a two-sample confirmation study.

Major depression is associated with altered static functional connectivity in various brain networks, particularly the default mode network (DMN). Dynamic functional connectivity is a novel tool with little application in affective disorders to date, and holds the potential to unravel fluctuations in connectivity strength over time in major depression. We assessed stability of connectivity in major depression between the medial prefrontal cortex (mPFC) and posterior cingulate cortex (PCC), key nodes in the DMN that are implicated in ruminative cognitions. Functional connectivity stability between the mPFC and PCC over the course of a resting-state functional magnetic resonance imaging (fMRI) scan was compared between medication-free patients with major depression and healthy controls matched for age, sex and handedness. We tested replicability of the results in an independent sample using multi-echo resting-state fMRI. The primary sample included 20 patients and 19 controls, while the validation sample included 19 patients and 19 controls. Greater connectivity variability was detected in major depression between mPFC and PCC. This was demonstrated in both samples indicating that the results were reliable and were not influenced by the fMRI acquisition approach used. Our results demonstrate that alterations within the DMN in major depression go beyond changes in connectivity strength and extend to reduced connectivity stability within key DMN regions. Findings were robustly replicated across two independent samples. Further research is necessary to better understand the nature of these fluctuations in connectivity and their relationship to the aetiology of major depression.

Ketamine induces a robust whole-brain connectivity pattern that can be differentially modulated by drugs of different mechanism and clinical profile.

Ketamine, an N-methyl-D-aspartate receptor (NMDAR) antagonist, has been studied in relation to the glutamate hypothesis of schizophrenia and increases dissociation, positive and negative symptom ratings. Ketamine effects brain function through changes in brain activity; these activity patterns can be modulated by pre-treatment of compounds known to attenuate the effects of ketamine on glutamate release. Ketamine also has marked effects on brain connectivity; we predicted that these changes would also be modulated by compounds known to attenuate glutamate release. Here, we perform task-free pharmacological magnetic resonance imaging (phMRI) to investigate the functional connectivity effects of ketamine in the brain and the potential modulation of these effects by pre-treatment of the compounds lamotrigine and risperidone, compounds hypothesised to differentially modulate glutamate release. Connectivity patterns were assessed by combining windowing, graph theory and multivariate Gaussian process classification. We demonstrate that ketamine has a robust effect on the functional connectivity of the human brain compared to saline (87.5 % accuracy). Ketamine produced a shift from a cortically centred, to a subcortically centred pattern of connections. This effect is strongly modulated by pre-treatment with risperidone (81.25 %) but not lamotrigine (43.75 %). Based on the differential effect of these compounds on ketamine response, we suggest the observed connectivity effects are primarily due to NMDAR blockade rather than downstream glutamatergic effects. The connectivity changes contrast with amplitude of response for which no differential effect between pre-treatments was detected, highlighting the necessity of these techniques in forming an informed view of the mechanistic effects of pharmacological compounds in the human brain.