Mon-Ju Wu

Hippocampal subfield volumes in mood disorders

Volume reduction and shape abnormality of the hippocampus have been associated with mood disorders. However, the hippocampus is not a uniform structure and consists of several subfields, such as the cornu ammonis (CA) subfields CA1–4, the dentate gyrus (DG) including a granule cell layer (GCL) and a molecular layer (ML) that continuously crosses adjacent subiculum (Sub) and CA fields. It is known that cellular and molecular mechanisms associated with mood disorders may be localized to specific hippocampal subfields. Thus, it is necessary to investigate the link between the in vivo hippocampal subfield volumes and specific mood disorders, such as bipolar disorder (BD) and major depressive disorder (MDD). In the present study, we used a state-of-the-art hippocampal segmentation approach, and we found that patients with BD had reduced volumes of hippocampal subfields, specifically in the left CA4, GCL, ML and both sides of the hippocampal tail, compared with healthy subjects and patients with MDD. The volume reduction was especially severe in patients with bipolar I disorder (BD-I). We also demonstrated that hippocampal subfield volume reduction was associated with the progression of the illness. For patients with BD-I, the volumes of the right CA1, ML and Sub decreased as the illness duration increased, and the volumes of both sides of the CA2/3, CA4 and hippocampal tail had negative correlations with the number of manic episodes. These results indicated that among the mood disorders the hippocampal subfields were more affected in BD-I compared with BD-II and MDD, and manic episodes had focused progressive effect on the CA2/3 and CA4 and hippocampal tail.

Changes in the corpus callosum in women with late-stage bipolar disorder.

This study investigated the differences in corpus callosum (CC) volumes between women with early‐stage and late‐stage bipolar I (BP I) disorder using the criteria previously described in the literature.

Brain gyrification and neuroprogression in bipolar disorder

Bipolar disorder has a prevalence of about 1%–5% worldwide and is associated with premature death by multiple causes, including cardiovascular disease, diabetes, and suicide. Less appreciated, however, is the emerging evidence suggesting that bipolar disorder may present a progressive course with neuroanatomical changes (1). In this sense, the term neuroprogression was put forward as the pathological rewiring of the brain that takes place in parallel with the cognitive and clinical deterioration in the course of bipolar disorder (1). Brain gyrification is an important anatomical characteristic of the human cortex. The spatial folding due to gyrification makes it possible for our brain to host more cortical neurons within a limited cranial volume than a brain without cortical gyrification. Some studies reported that the gyrification in patients with bipolar disorder was altered (2). However, the relationship between the brain gyrification and neuroprogression in bipolar disorder was still unknown. Previous studies on neuroprogression categorize the progressive stages of bipolar disorder according to prior numbers of manic episodes and hospitalizations, and provided valuable insights of the relationship between the brain changes and the neuroprogression of bipolar disorder (3, 4). In the current study, we used the same method to classify subjects with bipolar disorder as ‘BD-Late’, if they had 10 or more manic episodes and one or more hospitalizations due to manic or depressive episodes and as ‘BD-Early’, if they had three or less manic episodes. The remaining subjects were classified as the ‘Intermediate-stage’ (BD-Intermediate). Sixty-nine patients with bipolar I disorder (16 BD-Early, 38 BD-Intermediate and 15 BD-Late) according to DSM-IV and 80 healthy controls were recruited. Patients with head trauma with residual effects, neurological disorder, and uncontrolled major medical conditions were excluded. Axis-I diagnoses and clinical characteristics were assessed with the Structured Clinical Interview for DSM-IV axis-I Disorders (SCID-I). Patients with comorbidities were not excluded. However, only nine (13%) patients had comorbidities with substance abuse and 14 (20%) with PTSD, and their distribution in the three stage groups were not significantly different under v tests (P > 0.05). Only seven percent of patients were in mania, 46% patients were in depression and 22% were euthymic at the time of the scan. The mood states were not significantly different across the three stage groups. Current dimensional mood symptoms were assessed with the Hamilton Depression Scale (HAM-D) and the Young Mania Rating Scale (YMRS). All the subjects signed written consent forms and the study was approved by local IRB committee. We acquired structural T1-weighted scans using a Philips 1.5 Tesla MRI scanner (Philips Medical System, Andover, MA, USA) with a three-dimensional axial fast field echo sequence. The parameters were as reported previously (3). The average cortical gyrification of all subjects was the average of local gyrification index (GI) at each cortical surface vertex estimated using the Freesurfer software suite version 5.3.0 (http://surfer.nmr.mgh.harvard.edu). The GI is the ratio of the cortical surface area to an envelope surface that smoothly contains the brain (5). We used the general linear model and Spearman’s correlation to estimate the effect of stages (HC, BD-Early, BD-Intermediate, BD-Late; severity from low to high) on the cortical GI and local GI on the cortical surface. We considered Pvalues <0.05 significant. We found a significant stage effect on the GI (F (3,143) = 3.792; P = 0.050). Further post hoc analysis showed that BD-Intermediate (P = 0.034) and BD-Late (P = 0.025) subjects had significantly lower GI than HC, while BD-Early had similar GI with HC (P > 0.05), but these results did not survive Bonferroni correction. Furthermore, the correlation between the GI and stages of bipolar disorder was significant (r = 0.181, P = 0.027, confidence interval with bootstrapping [ 0.286, 0.077]). It is worth mentioning that illness duration was not different among groups. Surface-based analysis based on Spearman’s correlation also found negative local GI changes from early to late stages at a false discovery rate (FDR) of 0.3 across the cortical surface. These results indicate progressive changes of brain gyrification in different stages of bipolar disorder, and provide evidence of the brain gyrification as a marker of bipolar disorder stages. These findings are consistent with previous findings about the pathophysiologic associations between number of episodes and hospitalization, brain alterations and cognitive impairment, and further support the neuroprogression theory (1).

Deep Learning for Quality Control of Subcortical Brain 3D Shape Models

We present several deep learning models for assessing the morphometric fidelity of deep grey matter region models extracted from brain MRI. We test three different convolutional neural net architectures (VGGNet, ResNet and Inception) over 2D maps of geometric features. Further, we present a novel geometry feature augmentation technique based on parametric spherical mapping. Finally, we present an approach for model decision visualization, allowing human raters to see the areas of subcortical shapes most likely to be deemed of failing quality by the machine. Our training data is comprised of 5200 subjects from the ENIGMA Schizophrenia MRI cohorts, and our test dataset contains 1500 subjects from the ENIGMA Major Depressive Disorder cohorts. Our final models reduce human rater time by 46-70%. ResNet outperforms VGGNet and Inception for all of our predictive tasks.