Sensory, somatomotor and internal mentation networks emerge dynamically in the resting brain with internal mentation predominating in older age

Latent Similarity Identifies Important Functional Connections for Phenotype Prediction

OBJECTIVE

Endophenotypes such as brain age and fluid intelligence are important biomarkers of disease status. However, brain imaging studies to identify these biomarkers often encounter limited numbers of subjects but high dimensional imaging features, hindering reproducibility. Therefore, we develop an interpretable, multivariate classification/regression algorithm, called Latent Similarity (LatSim), suitable for small sample size but high feature dimension datasets.

METHODS

LatSim combines metric learning with a kernel similarity function and softmax aggregation to identify task-related similarities between subjects. Inter-subject similarity is utilized to improve performance on three prediction tasks using multi-paradigm fMRI data. A greedy selection algorithm, made possible by LatSim's computational efficiency, is developed as an interpretability method.

RESULTS

LatSim achieved significantly higher predictive accuracy at small sample sizes on the Philadelphia Neurodevelopmental Cohort (PNC) dataset. Connections identified by LatSim gave superior discriminative power compared to those identified by other methods. We identified 4 functional brain networks enriched in connections for predicting brain age, sex, and intelligence.

CONCLUSION

We find that most information for a predictive task comes from only a few (1-5) connections. Additionally, we find that the default mode network is over-represented in the top connections of all predictive tasks.

SIGNIFICANCE

We propose a novel prediction algorithm for small sample, high feature dimension datasets and use it to identify connections in task fMRI data. Our work can lead to new insights in both algorithm design and neuroscience research.

Age-dependent changes in the dynamic functional organization of the brain at rest: a cross-cultural replication approach

Age-associated changes in brain function play an important role in the development of neurodegenerative diseases. Although previous work has examined age-related changes in static functional connectivity, accumulating evidence suggests that advancing age is especially associated with alterations in the dynamic interactions and transitions between different brain states, which hitherto have received less attention. Conclusions of previous studies in this domain are moreover limited by suboptimal replicability of resting-state functional magnetic resonance imaging (fMRI) and culturally homogenous cohorts. Here, we investigate the robustness of age-associated changes in dynamic functional connectivity (dFC) by capitalizing on the availability of fMRI cohorts from two cultures (Western European and Chinese). In both the LEMON (Western European) and SALD (Chinese) cohorts, we consistently identify two distinct states: a more frequent segregated within-network connectivity state (state I) and a less frequent integrated between-network connectivity state (state II). Moreover, in both these cohorts, older (55-80 years) compared to younger participants (20-35 years) exhibited lower occurrence of and spent less time in state I. Older participants also tended to exhibit more transitions between networks and greater variance in global efficiency. Overall, our cross-cultural replication of age-associated changes in dFC metrics implies that advancing age is robustly associated with a reorganization of dynamic brain activation that favors the use of less functionally specific networks.

The potential shared brain functional alterations between adults with ADHD and children with ADHD co-occurred with disruptive behaviors

BACKGROUND

Attention-deficit/hyperactivity disorder (ADHD) is a common neurodevelopmental disorder. Many previous studies have shown that the comorbid status of disruptive behaviour disorders (DBD) was a predictor for ADHD persistence into adulthood. However, the brain mechanisms underlying such a relationship remain unclear. Thus, we aim to investigate whether the brain functional alteration in adults with ADHD could also be detected in children with ADHD co-occurring with disruptive behaviours from both quantitative and categorical dimensions.

METHODS

A total of 172 children with ADHD (cADHD), 98 adults with ADHD (aADHD), 77 healthy control children (cHC) and 40 healthy control adults (aHC) were recruited. The whole-brain spontaneous fluctuations in brain activity of each participant were recorded using functional near-infrared spectroscopy (fNIRS), and the functional connectivities (FCs) were calculated. We first compared the FC differences between aADHD and aHC. Then, for the regions with significantly abnormal FCs in aADHD, we further compared these features between cADHD and cHC. In addition, the correlation between these FCs and the conduct disorder (CD)/oppositional defiant disorder (ODD) symptoms were analysed in cADHD. Moreover, to render the results readily interpretable, we compared the FC differences among ADHD_CD-, subthreshold ADHD_CD+ and cHC groups, and among ADHD_ODD-, ADHD_ODD+ and cHC groups. Finally, we repeated the above analysis after controlling for other comorbidities and core symptoms to diminish the potential confounding effects.

RESULTS

We found that compared with aHC, aADHD showed significantly increased FCs in the VN, DMN, SMN, and DAN. The aforementioned abnormal FCs were also detected in cADHD, however, in an opposite orientation. Notably, these abnormal FCs were positively correlated with CD symptoms. Finally, the subthreshold ADHD_CD+ group even exhibited a tendency of adult-like increased FCs compared with the cHC. The results held after controlling for other comorbidities and core symptoms.

CONCLUSION

This study provides functional neuroimaging evidence that CD might be a risk factor for ADHD persistence into adulthood. Our work highlights the importance of differentiating ADHD_CD+ from ADHD and inspiring further understanding of brain development in ADHD.