Search

Using machine learning, we recently decomposed the neuroanatomical heterogeneity of established schizophrenia to discover two volumetric subgroups—a ‘lower brain volume’ subgroup (SG1) and an ‘higher striatal volume’ subgroup (SG2) with otherwise normal brain structure. In this study, we investigated whether the MRI signatures of these subgroups were also already present at the time of the first-episode of psychosis (FEP) and whether they were related to clinical presentation and clinical remission over 1-, 3-, and 5-years. We included 572 FEP and 424 healthy controls (HC) from 4 sites (Sao Paulo, Santander, London, Melbourne) of the PHENOM consortium. Our prior MRI subgrouping models (671 participants; USA, Germany, and China) were applied to both FEP and HC. Participants were assigned into 1 of 4 categories: subgroup 1 (SG1), subgroup 2 (SG2), no subgroup membership (‘None’), and mixed SG1 + SG2 subgroups (‘Mixed’). Voxel-wise analyses characterized SG1 and SG2 subgroups. Supervised machine learning analyses characterized baseline and remission signatures related to SG1 and SG2 membership. The two dominant patterns of ‘lower brain volume’ in SG1 and ‘higher striatal volume’ (with otherwise normal neuromorphology) in SG2 were identified already at the first episode of psychosis. SG1 had a significantly higher proportion of FEP (32%) vs. HC (19%) than SG2 (FEP, 21%; HC, 23%). Clinical multivariate signatures separated the SG1 and SG2 subgroups (balanced accuracy = 64%; p< 0.0001), with SG2 showing higher education but also greater positive psychosis symptoms at first presentation, and an association with symptom remission at 1-year, 5-year, and when timepoints were combined. Neuromorphological subtypes of schizophrenia are already evident at illness onset, separated by distinct clinical presentations, and differentially associated with subsequent remission. These results suggest that the subgroups may be underlying risk phenotypes that could be targeted in future treatment trials and are critical to consider when interpreting neuroimaging literature.

Human brain structure changes throughout the lifespan. Altered brain growth or rates of decline are implicated in a vast range of psychiatric, developmental and neurodegenerative diseases. In this study, we identified common genetic variants that affect rates of brain growth or atrophy in what is, to our knowledge, the first genome-wide association meta-analysis of changes in brain morphology across the lifespan. Longitudinal magnetic resonance imaging data from 15,640 individuals were used to compute rates of change for 15 brain structures. The most robustly identified genes GPR139, DACH1and APOEare associated with metabolic processes. We demonstrate global genetic overlap with depression, schizophrenia, cognitive functioning, insomnia, height, body mass index and smoking. Gene set findings implicate both early brain development and neurodegenerative processes in the rates of brain changes. Identifying variants involved in structural brain changes may help to determine biological pathways underlying optimal and dysfunctional brain development and aging.

Bipolar disorders (BDs) are among the leading causes of morbidity and disability. Objective biological markers, such as those based on brain imaging, could aid in clinical management of BD. Machine learning (ML) brings neuroimaging analyses to individual subject level and may potentially allow for their diagnostic use. However, fair and optimal application of ML requires large, multi-site datasets. We applied ML (support vector machines) to MRI data (regional cortical thickness, surface area, subcortical volumes) from 853 BD and 2167 control participants from 13 cohorts in the ENIGMA consortium. We attempted to differentiate BD from control participants, investigated different data handling strategies and studied the neuroimaging/clinical features most important for classification. Individual site accuracies ranged from 45.23% to 81.07%. Aggregate subject-level analyses yielded the highest accuracy (65.23%, 95% CI?=?63.47–67.00, ROC-AUC?=?71.49%, 95% CI?=?69.39–73.59), followed by leave-one-site-out cross-validation (accuracy = 58.67%, 95% CI?=?56.70–60.63). Meta-analysis of individual site accuracies did not provide above chance results. There was substantial agreement between the regions that contributed to identification of BD participants in the best performing site and in the aggregate dataset (Cohen’s Kappa?=?0.83, 95% CI?=?0.829–0.831). Treatment with anticonvulsants and age were associated with greater odds of correct classification. Although short of the 80% clinically relevant accuracy threshold, the results are promising and provide a fair and realistic estimate of classification performance, which can be achieved in a large, ecologically valid, multi-site sample of BD participants based on regional neurostructural measures. Furthermore, the significant classification in different samples was based on plausible and similar neuroanatomical features. Future multi-site studies should move towards sharing of raw/voxelwise neuroimaging data.

Psychiatric prognosis is a difficult problem. Making a prognosis requires looking far into the future, as opposed to making a diagnosis, which is concerned with the current state. During the follow-up period, many factors will influence the course of the disease. Combined with the usually scarcer longitudinal data and the variability in the definition of outcomes/transition, this makes prognostic predictions a challenging endeavor. Employing neuroimaging data in this endeavor introduces the additional hurdle of high dimensionality. Machine learning techniques are especially suited to tackle this challenging problem. This review starts with a brief introduction to machine learning in the context of its application to clinical neuroimaging data. We highlight a few issues that are especially relevant for prediction of outcome and transition using neuroimaging. We then review the literature that discusses the application of machine learning for this purpose. Critical examination of the studies and their results with respect to the relevant issues revealed the following: 1) there is growing evidence for the prognostic capability of machine learning–based models using neuroimaging; and 2) reported accuracies may be too optimistic owing to small sample sizes and the lack of independent test samples. Finally, we discuss options to improve the reliability of (prognostic) prediction models. These include new methodologies and multimodal modeling. Paramount, however, is our conclusion that future work will need to provide properly (cross-)validated accuracy estimates of models trained on sufficiently large datasets. Nevertheless, with the technological advances enabling acquisition of large databases of patients and healthy subjects, machine learning represents a powerful tool in the search for psychiatric biomarkers.

The influence of antipsychotic medication on brain morphology in schizophrenia may confound interpretation of brain changes over time. We aimed to assess the effect of discontinuation of atypical antipsychotic medication on change in brain volume in patients. Sixteen remitted, stable patients with first-episode schizophrenia, schizoaffective or schizophreniform disorder and 20 healthy controls were included. Two magnetic resonance imaging brain scans were obtained from all subjects with a 1-year interval. The patients either discontinued (n = 8) their atypical antipsychotic medication (olanzapine, risperidone, or quetiapine) or did not (n = 8) discontinue during the follow-up period. Intracranial volume and volumes of total brain, cerebral gray and white matter, cerebellum, third and lateral ventricle, nucleus caudatus, nucleus accumbens, and putamen were obtained. Multiple linear regression analyses were used to assess main effects for group (patient-control) and discontinuation (yes-no) for brain volume (change) while correcting for age, sex, and intracranial volume. Decrease in cerebral gray matter and caudate nucleus volume over time was significantly more pronounced in patients relative to controls. Our data suggest decreases in the nucleus accumbens and putamen volumes during the interval in patients who discontinued antipsychotic medication, whereas increases were found in patients who continued their antipsychotics. We confirmed earlier findings of excessive gray matter volume decrements in patients with schizophrenia compared with normal controls. We found evidence suggestive of decreasing volumes of the putamen and nucleus accumbens over time after discontinuation of medication. This might suggest that discontinuation reverses effects of atypical medication.

Objective: Sex hormones are not only involved in the formation of reproductive organs, but also induce sexually-dimorphic brain development and organization. Cross-sex hormone administration to transsexuals provides a unique possibility to study the effects of sex steroids on brain morphology in young adulthood.Methods: Magnetic resonance brain images were made prior to, and during, cross-sex hormone treatment to study the influence of anti-androgen + estrogen treatment on brain morphology in eight young adult male-to-female transsexual human subjects and of androgen treatment in six female-to-male transsexuals.Results: Compared with controls, anti-androgen + estrogen treatment decreased brain volumes of male-to-female subjects towards female proportions, while androgen treatment in female-to-male subjects increased total brain and hypothalamus volumes towards male proportions.Conclusions: The findings suggest that, throughout life, gonadal hormones remain essential for maintaining aspects of sex-specific differences in the human brain.

It has been suggested that genes involved in dopamine neurotransmission contribute to the pathogenesis of schizophrenia. However, reported associations of the disorder with genetic markers in dopaminergic genes have yielded inconsistent results. Possible explanations are differences in phenotyping, genetic heterogeneity, low marker informativity, and the use of small sample sizes. Here, we present a two‐stage analysis of 12 dopaminergic genes in a large sample of Dutch schizophrenic patients. To reduce genetic heterogeneity, only patients with at least three Caucasian grandparents of Dutch ancestry were ascertained. An efficient genotyping strategy was used, in which polymorphic microsatellite markers were first screened for association in DNA pools. Promising results were followed up by individual genotyping in an extended sample. The pooled samples consisted of 208 schizophrenic patients and 288 unmatched control individuals. For each of the genes, more than one microsatellite marker was selected where possible, either intragenic or close to the gene. After correcting for multiple testing, significantly different allele frequencies were detected for DRD5 marker D4S615. Subsequently, we individually genotyped this particular marker and another DRD5 marker, as well as a DRD3 marker that could not be analyzed using the pooling strategy. This was done in an extended sample of 282 schizophrenic patients and a control sample of 585 individuals. In this second stage of the study, we found no association between these three markers and schizophrenia. The results of our comprehensive analysis provide no evidence for association between schizophrenia and 12 dopaminergic genes in a large Dutch sample. © 2004 Wiley‐Liss, Inc.

Twin studies are important to investigate genetic influences on variation in human brain morphology in health and disease. However, the twin method has been criticized for its alleged non‐generalizability due to differences in the intrauterine and family environment of twins, compared with singletons. To test whether twin–singleton differences complicate interpretation of genetic contributions on variation in brain volume, brains from 112 pairs of twins and 34 of their siblings with a mean (standard deviation) age of 30.7 (9.6) years were scanned using MRI. The influence of birth order, zygosity and twin–sibling differences on brain volume measures was analysed using maximum‐likelihood model fitting. Variances were homogeneous across birth order, zygosity and twin–singleton status. Irrespective of zygosity, intracranial volume was smaller in second‐born twins compared with first‐born twins and compared with siblings. Grey matter volume was smaller in second‐born twins compared with first‐born twins. White matter was smaller in twins compared with siblings. Differences in grey and white matter between these groups were no longer significant after correction for intracranial volume. Total brain, and lateral and third ventricle volumes were comparable in twins and singletons. In conclusion, second‐born twins have a smaller intracranial volume than their first‐born co‐twins and siblings. This suggests aberrant early brain development in second‐born twins, which is consistent with the suboptimal pre‐ and perinatal environment related to birth order in twins. Since other brain volume measures were comparable between the groups, twin studies can provide reliable estimates of heritabilities in brain volume measures and these can be generalized to the singleton population.

Twin studies are important to investigate genetic influences on variation in human brain morphology in health and disease. However, the twin method has been criticized for its alleged non-generalizability due to differences in the intrauterine and family environment of twins, compared with singletons. To test whether twin-singleton differences complicate interpretation of genetic contributions on variation in brain volume, brains from 112 pairs of twins and 34 of their siblings with a mean (standard deviation) age of 30.7 (9.6) years were scanned using MRI. The influence of birth order, zygosity and twin-sibling differences on brain volume measures was analysed using maximum-likelihood model fitting. Variances were homogeneous across birth order, zygosity and twin-singleton status. Irrespective of zygosity, intracranial volume was smaller in second-born twins compared with first-born twins and compared with siblings. Grey matter volume was smaller in second-born twins compared with first-born twins. White matter was smaller in twins compared with siblings. Differences in grey and white matter between these groups were no longer significant after correction for intracranial volume. Total brain, and lateral and third ventricle volumes were comparable in twins and singletons. In conclusion, second-born twins have a smaller intracranial volume than their first-born co-twins and siblings. This suggests aberrant early brain development in second-born twins, which is consistent with the suboptimal pre- and perinatal environment related to birth order in twins. Since other brain volume measures were comparable between the groups, twin studies can provide reliable estimates of heritabilities in brain volume measures and these can be generalized to the singleton population.

The aim of this study was to investigate whether frontal lobe damage affects thalamic volume in humans. Ipsilateral and contralateral thalamic areas were measured in 0.5 T T1-weighted sagittal magnetic resonance images in 12 patients, first at the time of their surgery for relief of a unilateral frontal lobe brain tumor and at follow-up ∼2 years later. A 5 decrease in ipsilateral and 4.5 increase in contralateral thalamic area was found over time (F(1,11) = 6.15, p< 0.05). We conclude that unilateral frontal lobe damage results in a decrease in the ipsilateral thalamus and an increase in the contralateral thalamus in humans in vivo.The findings may have implications for the interpretation of the reported changes in thalamic volume in neuropsychiatric diseases.