Machine learning techniques are increasingly being used in making relevant predictions and inferences on
individual
subjects neuroimaging scan data. Previous studies have mostly focused on ...categorical discrimination of patients and matched healthy controls and more recently, on prediction of
individual
continuous variables such as clinical scores or age. However, these studies are greatly hampered by the large number of predictor variables (voxels) and low observations (subjects) also known as the
curse-of-dimensionality
or
small-n-large-p
problem. As a result, feature reduction techniques such as feature subset selection and dimensionality reduction are used to remove redundant predictor variables and experimental noise, a process which mitigates the
curse-of-dimensionality
and
small-n-large-p
effects. Feature reduction is an essential step before training a machine learning model to avoid
overfitting
and therefore improving model prediction
accuracy
and generalization ability. In this review, we discuss feature reduction techniques used with machine learning in neuroimaging studies.
There is still little knowledge of objective suicide risk stratification.
This study aims to develop models using machine-learning approaches to predict suicide attempt (1) among survey participants ...in a nationally representative sample and (2) among participants with lifetime major depressive episodes. We used a cohort called the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) that was conducted in two waves and included a nationally representative sample of the adult population in the United States. Wave 1 involved 43 093 respondents and wave 2 involved 34 653 completed face-to-face reinterviews with wave 1 participants. Predictor variables included clinical, stressful life events, and sociodemographic variables from wave 1; outcome included suicide attempt between wave 1 and wave 2.
The model built with elastic net regularization distinguished individuals who had attempted suicide from those who had not with an area under the ROC curve (AUC) of 0.89, balanced accuracy 81.86%, specificity 89.22%, and sensitivity 74.51% for the general population. For participants with lifetime major depressive episodes, AUC was 0.89, balanced accuracy 81.64%, specificity 85.86%, and sensitivity 77.42%. The most important predictor variables were a diagnosis of borderline personality disorder, post-traumatic stress disorder, and being of Asian descent for the model in all participants; and previous suicide attempt, borderline personality disorder, and overnight stay in hospital because of depressive symptoms for the model in participants with lifetime major depressive episodes. Random forest and artificial neural networks had similar performance.
Risk for suicide attempt can be estimated with high accuracy.
Diffusion tensor imaging has the potential to be used as a neuroimaging marker of natural ageing and assist in elucidating trajectories of cerebral maturation and ageing. In this study, we applied a ...multivariate technique relevance vector regression (RVR) to predict individual subject's age using whole brain fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD) and radial diffusivity (RD) from a cohort of 188 subjects aged 4–85years. High prediction accuracy as derived from Pearson correlation coefficient of actual versus predicted age (FA — r=0.870 p<0.0001; MD — r=0.896 p<0.0001; AD — r=0.895 p<0.0001; RD — r=0.899 p<0.0001) was achieved. Cerebral white-matter regions that contributed to these predictions include; corpus callosum, cingulum bundles, posterior longitudinal fasciculus and the cerebral peduncle. A post-hoc analysis of these regions showed that FA follows a nonlinear rational-quadratic trajectory across the lifespan peaking at approximately 21.8years. The MD, RD and AD volumes were particularly useful for making predictions using grey matter cerebral regions. These results suggest that diffusion tensor imaging measurements can reliably predict individual subject's age and demonstrate that FA cerebral maturation and ageing patterns follow a non-linear trajectory with a noteworthy peaking age. These data will contribute to the understanding of neurobiology of cerebral maturation and ageing. Most notably, from a neuropsychiatric perspective our results may allow differentiation of cerebral changes that may occur due to natural maturation and ageing, and those due to developmental or neuropsychiatric disorders.
•Machine-learning is used to predict age using whole-brain diffusion tensor images.•A cross-validation approach is used to separate training and testing datasets.•White matter follows a rational-quadratic trajectory peaking at 21.8years.•Diffusivity in grey-matter tissue increases with maturation and ageing.
Cortical gyrification of the brain represents the folding characteristic of the cerebral cortex. How the brain cortical gyrification changes from childhood to old age in healthy human subjects is ...still unclear. Additionally, studies have shown regional gyrification alterations in patients with major psychiatric disorders, such as major depressive disorder (MDD), bipolar disorder (BD), and schizophrenia (SCZ). However, whether the lifespan trajectory of gyrification over the brain is altered in patients diagnosed with major psychiatric disorders is still unknown. In this study, we investigated the trajectories of gyrification in three independent cohorts based on structural brain images of 881 subjects from age 4 to 83. We discovered that the trajectory of gyrification during normal development and aging was not linear and could be modeled with a logarithmic function. We also found that the gyrification trajectories of patients with MDD, BD and SCZ were deviated from the healthy one during adulthood, indicating altered aging in the brain of these patients.
Altered peripheral immune/inflammatory system and brain volumetric changes have been implicated in the pathophysiology of bipolar disorder (BD). This study aimed to evaluate how peripheral levels of ...cytokines are related to volumetric brain changes in euthymic patients with BD.
Euthymic patients with BD (n = 21) and healthy controls (n = 22) were enrolled in this exploratory study. Blood samples were collected on the same day of clinical assessment and neuroimaging. Cytokines were measured through cytometric bead array method. Neuroimaging data were acquired using a sagittal three-dimensional magnetic resonance imaging T1-weighted fast field echo sequence and was processed using FreeSurfer.
Compared to controls, BD patients had significantly lower volumes in the cingulate, medial-orbitofrontal (MOF) and parahippocampal regions. We found a negative correlation between right MOF volume and interferon-gamma levels (
= -0.431,
= .049) and a positive correlation between interleukin-10 levels and left posterior cingulate volume (
= 0.457,
= .048).
Our results support the involvement of inflammatory pathways in structural brain changes in BD.
Abstract Background: The neuroanatomical substrate underlying Major Depressive Disorder (MDD) is incompletely understood. Recent reports have implicated the insular cortex. Methods: Two cohorts of ...participants with MDD were tested. In the first MDD cohort, we used standardised facial expression recognition tasks. In the second cohort, we focused on facial disgust recognition, a function associated with the insular cortex. T1 weighted MR imaging was used in the second cohort to test the hypothesis of abnormal insular volume being associated with impaired disgust recognition. Results: Disgust recognition was particularly impaired in both cohorts. In the second cohort, the magnitude of the disgust recognition deficit correlated with reduced insula grey matter volume. Exploring the idea of insula involvement in MDD further, we identified the insular cortex and the anterior cingulate cortex as key neural correlates of core symptoms, in that scores of 3 clinical scales (the Beck Depression Inventory, the Hamilton Depression Rating Scale, and the Snaith–Hamilton Pleasure Scale) correlated with grey matter volume in these structures. Limitations: MDD participants were clinically representative of specialist and academic psychiatric practice in the UK and presented with robust primary diagnoses; we did not exclude common co-morbidities such as anxiety and personality disorders. Conclusions: We propose that cognitive and emotional functions assumed to be associated with the insula are adversely affected in patients with MDD and that this may, therefore, represent the substrate for some core clinical features of MDD. Further exploration of the involvement of the insular cortex in MDD is warranted.
Objectives
The International Society for Bipolar Disorders Big Data Task Force assembled leading researchers in the field of bipolar disorder (BD), machine learning, and big data with extensive ...experience to evaluate the rationale of machine learning and big data analytics strategies for BD.
Method
A task force was convened to examine and integrate findings from the scientific literature related to machine learning and big data based studies to clarify terminology and to describe challenges and potential applications in the field of BD. We also systematically searched PubMed, Embase, and Web of Science for articles published up to January 2019 that used machine learning in BD.
Results
The results suggested that big data analytics has the potential to provide risk calculators to aid in treatment decisions and predict clinical prognosis, including suicidality, for individual patients. This approach can advance diagnosis by enabling discovery of more relevant data‐driven phenotypes, as well as by predicting transition to the disorder in high‐risk unaffected subjects. We also discuss the most frequent challenges that big data analytics applications can face, such as heterogeneity, lack of external validation and replication of some studies, cost and non‐stationary distribution of the data, and lack of appropriate funding.
Conclusion
Machine learning‐based studies, including atheoretical data‐driven big data approaches, provide an opportunity to more accurately detect those who are at risk, parse‐relevant phenotypes as well as inform treatment selection and prognosis. However, several methodological challenges need to be addressed in order to translate research findings to clinical settings.