Sridhar Kandala

Functional neuroimaging of high-risk 6-month-old infants predicts a diagnosis of autism at 24 months of age

Functional brain imaging of 6-month-old infants with a high familial risk for autism predicts a diagnosis of autism at 24 months of age. Predicting the future with brain imaging In a new study, Emerson et al. show that brain function in infancy can be used to accurately predict which high-risk infants will later receive an autism diagnosis. Using machine learning techniques that identify patterns in the brain’s functional connections, Emerson and colleagues were able to predict with greater than 96% accuracy whether a 6-month-old infant would develop autism at 24 months of age. These findings must be replicated, but they represent an important step toward the early identification of individuals with autism before its characteristic symptoms develop. Autism spectrum disorder (ASD) is a neurodevelopmental disorder characterized by social deficits and repetitive behaviors that typically emerge by 24 months of age. To develop effective early interventions that can potentially ameliorate the defining deficits of ASD and improve long-term outcomes, early detection is essential. Using prospective neuroimaging of 59 6-month-old infants with a high familial risk for ASD, we show that functional connectivity magnetic resonance imaging correctly identified which individual children would receive a research clinical best-estimate diagnosis of ASD at 24 months of age. Functional brain connections were defined in 6-month-old infants that correlated with 24-month scores on measures of social behavior, language, motor development, and repetitive behavior, which are all features common to the diagnosis of ASD. A fully cross-validated machine learning algorithm applied at age 6 months had a positive predictive value of 100% [95% confidence interval (CI), 62.9 to 100], correctly predicting 9 of 11 infants who received a diagnosis of ASD at 24 months (sensitivity, 81.8%; 95% CI, 47.8 to 96.8). All 48 6-month-old infants who were not diagnosed with ASD were correctly classified [specificity, 100% (95% CI, 90.8 to 100); negative predictive value, 96.0% (95% CI, 85.1 to 99.3)]. These findings have clinical implications for early risk assessment and the feasibility of developing early preventative interventions for ASD.

Accurate age classification of 6 and 12 month-old infants based on resting-state functional connectivity magnetic resonance imaging data

Highlights • SVMs classified 6 versus 12 month-old infants above chance based on fcMRI data alone.• We carefully accounted for the effects of fcMRI motion artifact.• These results coincide with a period of dramatic change in infant development.• Two interpretations about connections supporting this age categorization are given.

Joint Attention and Brain Functional Connectivity in Infants and Toddlers

Abstract Initiating joint attention (IJA), the behavioral instigation of coordinated focus of 2 people on an object, emerges over the first 2 years of life and supports social‐communicative functioning related to the healthy development of aspects of language, empathy, and theory of mind. Deficits in IJA provide strong early indicators for autism spectrum disorder, and therapies targeting joint attention have shown tremendous promise. However, the brain systems underlying IJA in early childhood are poorly understood, due in part to significant methodological challenges in imaging localized brain function that supports social behaviors during the first 2 years of life. Herein, we show that the functional organization of the brain is intimately related to the emergence of IJA using functional connectivity magnetic resonance imaging and dimensional behavioral assessments in a large semilongitudinal cohort of infants and toddlers. In particular, though functional connections spanning the brain are involved in IJA, the strongest brain‐behavior associations cluster within connections between a small subset of functional brain networks; namely between the visual network and dorsal attention network and between the visual network and posterior cingulate aspects of the default mode network. These observations mark the earliest known description of how functional brain systems underlie a burgeoning fundamental social behavior, may help improve the design of targeted therapies for neurodevelopmental disorders, and, more generally, elucidate physiological mechanisms essential to healthy social behavior development.

Transdiagnostic Associations Between Functional Brain Network Integrity and Cognition

Importance Cognitive impairment occurs across the psychosis spectrum and is associated with functional outcome. However, it is unknown whether these shared manifestations of cognitive dysfunction across diagnostic categories also reflect shared neurobiological mechanisms or whether the source of impairment differs. Objective To examine whether the general cognitive deficit observed across psychotic disorders is similarly associated with functional integrity of 2 brain networks widely implicated in supporting many cognitive domains. Design, Setting, and Participants A total of 201 healthy control participants and 375 patients with psychotic disorders from the Bipolar-Schizophrenia Network on Intermediate Phenotypes (B-SNIP) consortium were studied from September 29, 2007, to May 31, 2011. The B-SNIP recruited healthy controls and stable outpatients from 6 sites: Baltimore, Maryland; Boston, Massachusetts; Chicago, Illinois; Dallas, Texas; Detroit, Michigan; and Hartford, Connecticut. All participants underwent cognitive testing and resting-state functional magnetic resonance imaging. Data analysis was performed from April 28, 2015, to February 21, 2017. Main Outcomes and Measures The Brief Assessment of Cognition in Schizophrenia was used to measure cognitive ability. A principal axis factor analysis on the Brief Assessment of Cognition in Schizophrenia battery yielded a single factor (54% variance explained) that served as the measure of general cognitive ability. Functional network integrity measures included global and local efficiency of the whole brain, cingulo-opercular network (CON), frontoparietal network, and auditory network and exploratory analyses of all networks from the Power atlas. Group differences in network measures, associations between cognition and network measures, and mediation models were tested. Results The final sample for the current study included 201 healthy controls, 143 patients with schizophrenia, 103 patients with schizoaffective disorder, and 129 patients with psychotic bipolar disorder (mean [SD] age, 35.1 [12.0] years; 281 male [48.8%] and 295 female [51.2%]; 181 white [31.4%], 348 black [60.4%], and 47 other [8.2%]). Patients with schizophrenia (Cohen d = 0.36, P < .001) and psychotic bipolar disorder (Cohen d = 0.33, P = .002) had significantly reduced CON global efficiency compared with healthy controls. All patients with psychotic disorders had significantly reduced CON local efficiency, but the clinical groups did not differ from one another. The CON global efficiency was significantly associated with general cognitive ability across all groups (&bgr; = 0.099, P = .009) and significantly mediated the association between psychotic disorder status and general cognition (&bgr; = −0.037; 95% CI, −0.076 to −0.014). Subcortical network global efficiency was also significantly reduced in psychotic disorders (F3,587 = 4.01, P = .008) and positively predicted cognitive ability (&bgr; = 0.094, P = .009). Conclusions and Relevance These findings provide evidence that reduced CON and subcortical network efficiency play a role in the general cognitive deficit observed across the psychosis spectrum. They provide new support for the dimensional hypothesis that a shared neurobiological mechanism underlies cognitive impairment in psychotic disorders.

Tract-based analysis of white matter integrity in psychotic and nonpsychotic bipolar disorder

BACKGROUND At least 50% of individuals with bipolar disorder (BD) present with psychosis during their lifetime. Psychotic symptoms have sometimes been linked to specific genetic and phenotypic markers. This study aims to explore potential differences between bipolar disorder subtypes by measuring white matter integrity of the brain and relationships with clinical measures. METHODS Diffusion tensor imaging and clinical measures were acquired from 102 participants, grouped as psychotic bipolar disorder (PBD) (n=48), non-psychotic bipolar disorder (NBD) (n=24), and healthy controls (n=30). We utilized a powerful, automated tool (TRACULA: Tracts Constrained by Underlying Anatomy) to analyze the fractional anisotropy (FA) and mean diffusivity (MD) of 18 white matter tracts. RESULTS Decreased FA in numerous tracts was observed in bipolar disorder groups compared to healthy controls: bilateral cingulum-cingulate gyrus bundles, corticospinal tracts, and superior longitudinal fasciculi as well as the right hemisphere cingulum-angular bundle. Only left uncinate fasciculus FA differed between PBD and NPBD groups. We found no group differences in MD. Positive symptoms correlated with FA in the superior (inversely) and inferior (directly) longitudinal fasciculi. Negative symptoms directly correlated with mean FA of the corticospinal tract and cingulum-angular bundle. LIMITATIONS Neurotropic, mood-stabilizing medication prescribed for individuals with BD may interact with measures of white matter integrity in our BD participants. CONCLUSION Our results indicate decreased white matter coherence in BD. Minimal differences in white matter FA between PBD and NPBD participants suggest related underlying neurobiology.

The Lifespan Human Connectome Project in Development: A large-scale study of brain connectivity development in 5–21 year olds

&NA; Recent technological and analytical progress in brain imaging has enabled the examination of brain organization and connectivity at unprecedented levels of detail. The Human Connectome Project in Development (HCP‐D) is exploiting these tools to chart developmental changes in brain connectivity. When complete, the HCP‐D will comprise approximately ˜1750 open access datasets from 1300 + healthy human participants, ages 5–21 years, acquired at four sites across the USA. The participants are from diverse geographical, ethnic, and socioeconomic backgrounds. While most participants are tested once, others take part in a three‐wave longitudinal component focused on the pubertal period (ages 9–17 years). Brain imaging sessions are acquired on a 3 T Siemens Prisma platform and include structural, functional (resting state and task‐based), diffusion, and perfusion imaging, physiological monitoring, and a battery of cognitive tasks and self‐reports. For minors, parents additionally complete a battery of instruments to characterize cognitive and emotional development, and environmental variables relevant to development. Participants provide biological samples of blood, saliva, and hair, enabling assays of pubertal hormones, health markers, and banked DNA samples. This paper outlines the overarching aims of the project, the approach taken to acquire maximally informative data while minimizing participant burden, preliminary analyses, and discussion of the intended uses and limitations of the dataset.

Rapid instruction-based task learning (RITL) in schizophrenia.

Individuals with schizophrenia demonstrate broad impairments in neurocognitive functioning as measured through laboratory-based tasks. Neuropsychological measures depend on rapid instruction-based task learning (RITL), the ability to rapidly translate task instruction into goal-directed behavior. Here, the authors present the first known investigation of RITL in schizophrenia and aim to test whether RITL deficits exist in schizophrenia, are associated with abnormal brain activation, and contribute to the generalized cognitive deficit. Twenty-nine schizophrenia participants and 31 healthy controls completed a previously established RITL task while in a functional magnetic resonance imaging (fMRI) scanner and completed a brief assessment of general cognition outside the scanner. Patients were significantly impaired in RITL accuracy and reaction time (RT). Compared to controls, patients had reduced activation of the caudate and left inferior frontal junction (LIFJ) while viewing task instructions, and across all subjects, lower activation in these regions was associated with worse RITL performance. During practice trials, activation in the anterior insula, LIFJ, and middle frontal gyrus also related to performance. RITL ability was robustly associated with general cognitive ability, explained a significant proportion of the variance in the generalized cognitive deficit, and was associated with LIFJ activity during RITL instructions. These results indicate that the ability to rapidly learn task instructions is impaired in schizophrenia and associated with abnormal activation of the caudate and LIFJ. Abnormalities in RITL represent a critical cognitive facet for understanding the broad profile of cognitive deficits in schizophrenia.

Multimodal neural correlates of cognitive control in the Human Connectome Project

&NA; Cognitive control is a construct that refers to the set of functions that enable decision‐making and task performance through the representation of task states, goals, and rules. The neural correlates of cognitive control have been studied in humans using a wide variety of neuroimaging modalities, including structural MRI, resting‐state fMRI, and task‐based fMRI. The results from each of these modalities independently have implicated the involvement of a number of brain regions in cognitive control, including dorsal prefrontal cortex, and frontal parietal and cingulo‐opercular brain networks. However, it is not clear how the results from a single modality relate to results in other modalities. Recent developments in multimodal image analysis methods provide an avenue for answering such questions and could yield more integrated models of the neural correlates of cognitive control. In this study, we used multiset canonical correlation analysis with joint independent component analysis (mCCA + jICA) to identify multimodal patterns of variation related to cognitive control. We used two independent cohorts of participants from the Human Connectome Project, each of which had data from four imaging modalities. We replicated the findings from the first cohort in the second cohort using both independent and predictive analyses. The independent analyses identified a component in each cohort that was highly similar to the other and significantly correlated with cognitive control performance. The replication by prediction analyses identified two independent components that were significantly correlated with cognitive control performance in the first cohort and significantly predictive of performance in the second cohort. These components identified positive relationships across the modalities in neural regions related to both dynamic and stable aspects of task control, including regions in both the frontal‐parietal and cingulo‐opercular networks, as well as regions hypothesized to be modulated by cognitive control signaling, such as visual cortex. Taken together, these results illustrate the potential utility of multi‐modal analyses in identifying the neural correlates of cognitive control across different indicators of brain structure and function. HighlightsIdentified independent components (ICs) that link across multiple types of imaging.Two ICs significantly correlated with cognitive control performance.Results were replicated using predictive and independent analyses in two cohorts.ICs included regions thought to be related to stable and dynamic cognitive control.

Functional connectivity of cognitive brain networks in schizophrenia during a working memory task

Task-based connectivity studies facilitate the understanding of how the brain functions during cognition, which is commonly impaired in schizophrenia (SZ). Our aim was to investigate functional connectivity during a working memory task in SZ. We hypothesized that the task-negative (default mode) network and the cognitive control (frontoparietal) network would show dysconnectivity. Twenty-five SZ patient and 31 healthy control scans were collected using the customized 3T Siemens Skyra MRI scanner, previously used to collect data for the Human Connectome Project. Blood oxygen level dependent signal during the 0-back and 2-back conditions were extracted within a network-based parcelation scheme. Average functional connectivity was assessed within five brain networks: frontoparietal (FPN), default mode (DMN), cingulo-opercular (CON), dorsal attention (DAN), and ventral attention network; as well as between the DMN or FPN and other networks. For within-FPN connectivity, there was a significant interaction between n-back condition and group (p = 0.015), with decreased connectivity at 0-back in SZ subjects compared to controls. FPN-to-DMN connectivity also showed a significant condition × group effect (p = 0.003), with decreased connectivity at 0-back in SZ. Across groups, connectivity within the CON and DAN were increased during the 2-back condition, while DMN connectivity with either CON or DAN were decreased during the 2-back condition. Our findings support the role of the FPN, CON, and DAN in working memory and indicate that the pattern of FPN functional connectivity differs between SZ patients and control subjects during the course of a working memory task.

Transdiagnostic multimodal neuroimaging in psychosis: structural, resting-state, and task MRI correlates of cognitive control

Background Psychotic disorders, including schizophrenia and bipolar disorder, are associated with impairments in regulation of goal-directed behavior, termed cognitive control. Cognitive control related neural alterations have been studied in psychosis. However, studies are typically unimodal and relationships across modalities of brain function and structure remain unclear. Thus, we performed transdiagnostic multimodal analyses to examine cognitive control related neural variation in psychosis. Methods Structural, resting, and working memory task imaging and behavioral data for 31 controls, 27 bipolar, and 23 schizophrenia patients were collected and processed identically to the Human Connectome Project (HCP), enabling identification of relationships with prior multimodal work. Two cognitive control related independent components (ICs) derived from the HCP using multiset canonical correlation analysis + joint independent component analysis (mCCA+jICA) were used to predict performance in psychosis. de novo mCCA+jICA was performed, and resultant IC weights were correlated with cognitive control. Results A priori ICs significantly predicted cognitive control in psychosis (3/5 modalities significant). De novo mCCA+jICA identified an IC correlated with cognitive control that also discriminated groups. Structural contributions included insular, somatomotor, cingulate, and visual regions; task contributions included precentral, posterior parietal, cingulate, and visual regions; and resting-state contributions highlighted canonical network organization. Follow-up analyses suggested de novo correlations with cognitive control were primarily influenced by schizophrenia patients. Conclusions A priori components partially predicted performance in transdiagnostic psychosis and de novo analyses identified novel contributions in somatomotor and visual regions in schizophrenia. Together, results suggest joint contributions across modalities related to cognitive control across the healthy-to-psychosis spectrum.