Search

Machine learning can be used to define subtypes of psychiatric conditions based on shared biological foundations of mental disorders. Here we analyzed cross-sectional brain images from 4,222 individuals with schizophrenia and 7038 healthy subjects pooled across 41 international cohorts from the ENIGMA, non-ENIGMA cohorts and public datasets. Using the Subtype and Stage Inference (SuStaIn) algorithm, we identify two distinct neurostructural subgroups by mapping the spatial and temporal trajectory of gray matter change in schizophrenia. Subgroup 1 was characterized by an early cortical-predominant loss with enlarged striatum, whereas subgroup 2 displayed an early subcortical-predominant loss in the hippocampus, striatum and other subcortical regions. We confirmed the reproducibility of the two neurostructural subtypes across various sample sites, including Europe, North America and East Asia. This imaging-based taxonomy holds the potential to identify individuals with shared neurobiological attributes, thereby suggesting the viability of redefining existing disorder constructs based on biological factors.

Machine learning can be used to define subtypes of psychiatric conditions based on shared clinical and biological foundations, presenting a crucial step toward establishing biologically based subtypes of mental disorders. With the goal of identifying subtypes of disease progression in schizophrenia, here we analyzed cross-sectional brain structural magnetic resonance imaging (MRI) data from 4,291 individuals with schizophrenia (1,709 females, age=32.5 years±11.9) and 7,078 healthy controls (3,461 females, age=33.0 years±12.7) pooled across 41 international cohorts from the ENIGMA Schizophrenia Working Group, non-ENIGMA cohorts and public datasets. Using a machine learning approach known as Subtype and Stage Inference (SuStaIn), we implemented a brain imaging-driven classification that identifies two distinct neurostructural subgroups by mapping the spatial and temporal trajectory of gray matter (GM) loss in schizophrenia. Subgroup 1 (n=2,622) was characterized by an early cortical-predominant loss (ECL) with enlarged striatum, whereas subgroup 2 (n=1,600) displayed an early subcortical-predominant loss (ESL) in the hippocampus, amygdala, thalamus, brain stem and striatum. These reconstructed trajectories suggest that the GM volume reduction originates in the Broca's area/adjacent fronto-insular cortex for ECL and in the hippocampus/adjacent medial temporal structures for ESL. With longer disease duration, the ECL subtype exhibited a gradual worsening of negative symptoms and depression/anxiety, and less of a decline in positive symptoms. We confirmed the reproducibility of these imaging-based subtypes across various sample sites, independent of macroeconomic and ethnic factors that differed across these geographic locations, which include Europe, North America and East Asia. These findings underscore the presence of distinct pathobiological foundations underlying schizophrenia. This new imaging-based taxonomy holds the potential to identify a more homogeneous sub-population of individuals with shared neurobiological attributes, thereby suggesting the viability of redefining existing disorder constructs based on biological factors.

Neuroanatomical abnormalities have been reported along a continuum from at-risk stages, including high schizotypy, to early and chronic psychosis. However, a comprehensive neuroanatomical mapping of schizotypy remains to be established. The authors conducted the first large-scale meta-analyses of cortical and subcortical morphometric patterns of schizotypy in healthy individuals, and compared these patterns with neuroanatomical abnormalities observed in major psychiatric disorders. The sample comprised 3004 unmedicated healthy individuals (12-68 years, 46.5% male) from 29 cohorts of the worldwide ENIGMA Schizotypy working group. Cortical and subcortical effect size maps with schizotypy scores were generated using standardized methods. Pattern similarities were assessed between the schizotypy-related cortical and subcortical maps and effect size maps from comparisons of schizophrenia (SZ), bipolar disorder (BD) and major depression (MDD) patients with controls. Thicker right medial orbitofrontal/ventromedial prefrontal cortex (mOFC/vmPFC) was associated with higher schizotypy scores (r = 0.067, pFDR = 0.02). The cortical thickness profile in schizotypy was positively correlated with cortical abnormalities in SZ (r = 0.285, pspin = 0.024), but not BD (r = 0.166, pspin = 0.205) or MDD (r = -0.274, pspin = 0.073). The schizotypy-related subcortical volume pattern was negatively correlated with subcortical abnormalities in SZ (rho = -0.690, pspin = 0.006), BD (rho = -0.672, pspin = 0.009), and MDD (rho = -0.692, pspin = 0.004). Comprehensive mapping of schizotypy-related brain morphometry in the general population revealed a significant relationship between higher schizotypy and thicker mOFC/vmPFC, in the absence of confounding effects due to antipsychotic medication or disease chronicity. The cortical pattern similarity between schizotypy and schizophrenia yields new insights into a dimensional neurobiological continuity across the extended psychosis phenotype.

Attention-deficit hyperactivity disorder (ADHD) is a debilitating disorder with apparent roots in abnormal brain development. Here, we quantified the level of individual brain maturation in children with ADHD using structural neuroimaging and a recently developed machine learning algorithm. More specifically, we compared the BrainAGE index between three groups matched for chronological age (mean ± SD: 11.86 ± 3.25 years): 89 children diagnosed with ADHD, 34 asymptomatic siblings of those children with ADHD, and 21 unrelated healthy control children. Brains of children with ADHD were estimated significantly younger (-0.85 years) than brains of healthy controls (Cohens d = -0.33; p = 0.028, one-tailed), while there were no significant differences between unaffected siblings and healthy controls. In addition, more severe ADHD symptoms were significantly associated with younger appearing brains. Altogether, these results are in line with the proposed delay of individual brain maturation in children with ADHD. However, given the relatively small sample size (N = 144), the findings should be considered preliminary and need to be confirmed in future studies.

Background: Anorexia nervosa is a psychiatric disorder characterised by undernutrition, significantly low body weight and large, although possibly transient, reductions in brain structure. Advanced brain ageing tracks accelerated age-related changes in brain morphology that have been linked to psychopathology and adverse clinical outcomes. Aim: The aim of the current case–control study was to characterise cross-sectional and longitudinal patterns of advanced brain age in acute anorexia nervosa and during the recovery process. Method: Measures of grey- and white-matter-based brain age were obtained from T1-weighted magnetic resonance imaging scans of 129 acutely underweight female anorexia nervosa patients (of which 95 were assessed both at baseline and after approximately 3 months of nutritional therapy), 39 recovered patients and 167 healthy female controls, aged 12–23 years. The difference between chronological age and grey- or white-matter-based brain age was calculated to indicate brain-predicted age difference (BrainAGEGM and BrainAGEWM). Results: Acute anorexia nervosa patients at baseline, but not recovered patients, showed a higher BrainAGEGM of 1.79 years (95% CI [1.45, 2.13]) compared to healthy controls. However, the difference was largely reduced for BrainAGEWM. After partial weight restoration, BrainAGEGM decreased substantially (beta = −1.69; CI [−1.93, −1.46]). BrainAGEs were unrelated to symptom severity or depression, but larger weight gain predicted larger normalisation of BrainAGEGM in the longitudinal patient sample (beta = −0.65; CI [−0.75, −0.54]). Conclusions: Our findings suggest that in patients with anorexia nervosa, undernutrition is an important predictor of advanced grey-matter-based brain age, which itself might be transient in nature and largely undetectable after weight recovery. [ABSTRACT FROM AUTHOR]

Congenital Adrenal Hyperplasia (CAH) has been reported to involve structural alterations in some brain regions. However, it remains to be established whether there is also an impact on the size of the brain as a whole. Here, we compiled the largest CAH sample to date (n = 53), matched pair-wise to a control group (n = 53) on sex, age, and verbal intelligence. Using T1-weighted brain scans, we calculated intracranial volume (ICV) as well as total brain volume (TBV), which are both common estimates for brain size. The statistical analysis was performed using a general linear model assessing the effects of CAH (CAH vs. controls), sex (women vs. men), and any CAH-by-sex interaction. The outcomes were comparable for ICV and TBV, i.e., there was no significant main effect of CAH and no significant CAH-by-sex interaction. However, there was a significant main effect of sex, with larger ICVs and TBVs in men than in women. Our findings contribute to an understudied field of research exploring brain anatomy in CAH. In contrast to some existing studies suggesting a smaller brain size in CAH, we did not observe such an effect. In other words, ICV and TBV in women and men with CAH did not differ significantly from those in controls. Notwithstanding, we observed the well-known sex difference in brain size (12.69% for ICV and 12.50% for TBV), with larger volumes in men than in women, which is in agreement with the existing literature. [ABSTRACT FROM AUTHOR]

Mild cognitive impairment (MCI) is a transitional stage between age-related cognitive decline and Alzheimer's disease (AD). For the effective treatment of AD, it would be important to identify MCI patients at high risk for conversion to AD. In this study, we present a novel magnetic resonance imaging (MRI)-based method for predicting the MCI-to-AD conversion from one to three years before the clinical diagnosis. First, we developed a novel MRI biomarker of MCI-to-AD conversion using semi-supervised learning and then integrated it with age and cognitive measures about the subjects using a supervised learning algorithm resulting in what we call the aggregate biomarker. The novel characteristics of the methods for learning the biomarkers are as follows: 1) We used a semi-supervised learning method (low density separation) for the construction of MRI biomarker as opposed to more typical supervised methods; 2) We performed a feature selection on MRI data from AD subjects and normal controls without using data from MCI subjects via regularized logistic regression; 3) We removed the aging effects from the MRI data before the classifier training to prevent possible confounding between AD and age related atrophies; and 4) We constructed the aggregate biomarker by first learning a separate MRI biomarker and then combining it with age and cognitive measures about the MCI subjects at the baseline by applying a random forest classifier. We experimentally demonstrated the added value of these novel characteristics in predicting the MCI-to-AD conversion on data obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. With the ADNI data, the MRI biomarker achieved a 10-fold cross-validated area under the receiver operating characteristic curve (AUC) of 0.7661 in discriminating progressive MCI patients (pMCI) from stable MCI patients (sMCI). Our aggregate biomarker based on MRI data together with baseline cognitive measurements and age achieved a 10-fold cross-validated AUC score of 0.9020 in discriminating pMCI from sMCI. The results presented in this study demonstrate the potential of the suggested approach for early AD diagnosis and an important role of MRI in the MCI-to-AD conversion prediction. However, it is evident based on our results that combining MRI data with cognitive test results improved the accuracy of the MCI-to-AD conversion prediction.

The interrelation between acute ischemic stroke, persistent disability, and uncertain prognosis underscores the need for improved methods to predict clinical outcomes. Traditional approaches have largely focused on analysis of clinical metrics, lesion characteristics, and network connectivity, using techniques such as resting-state functional magnetic resonance imaging (rs-fMRI) and diffusion tensor imaging (DTI). However, these methods are not routinely used in acute stroke diagnostics. This study introduces an innovative approach that not only considers the lesion size in relation to the National Institutes of Health Stroke Scale (NIHSS score), but also evaluates the impact of disrupted fibers and their connections to cortical regions by introducing a disconnection value. By identifying fibers traversing the lesion and estimating their number within predefined regions of interest (ROIs) using a normative connectome atlas, our method bypasses the need for individual DTI scans. In our analysis of MRI data (T1 and T2) from 51 patients with acute or subacute subcortical stroke presenting with motor or sensory deficits, we used simple linear regression to assess the explanatory power of lesion size and disconnection value on NIHSS score. Subsequent hierarchical multiple linear regression analysis determined the incremental value of disconnection metrics over lesion size alone in relation to NIHSS score. Our results showed that models incorporating the disconnection value accounted for more variance than those based solely on lesion size (lesion size explained 44% variance, disconnection value 60%). Furthermore, hierarchical regression revealed a significant improvement (p < 0.001) in model fit when adding the disconnection value, confirming its critical role in stroke assessment. Our approach, which integrates a normative connectome to quantify disconnections to cortical regions, provides a significant improvement in assessing the current state of stroke impact compared to traditional measures that focus on lesion size. This is achieved by taking into account the lesion's location and the connectivity of the affected white matter tracts, providing a more comprehensive assessment of stroke severity as reflected in the NIHSS score. Future research should extend the validation of this approach to larger and more diverse populations, with a focus on refining its applicability to clinical assessment and long-term outcome prediction. [ABSTRACT FROM AUTHOR]