The functional and structural connectomes associated with geriatric depression and anxiety symptoms in mild cognitive impairment: Cross-syndrome overlap and generalization

Individualized prediction of cognitive test scores from functional brain connectome in patients with first-episode late-life depression

BACKGROUND

In the realm of cognitive screening, the Mini-Mental State Examination (MMSE) and the Montreal Cognitive Assessment (MoCA) are widely utilized for detecting cognitive deficits in patients with late-life depression (LLD), However, the interindividual variability in neuroimaging biomarkers contributing to individual-specific symptom severity remains poorly understood. In this study, we used a connectome-based predictive model (CPM) approach on resting-state functional magnetic resonance imaging data from patients with LLD to establish individualized prediction models for the MoCA and the MMSE scores.

METHODS

We recruited 135 individuals diagnosed with first-episode LLD for this research. Participants underwent the MMSE and MoCA tests, along with resting-state functional magnetic resonance imaging scans. Functional connectivity matrices derived from these scans were utilized in CPM models to predict MMSE or MoCA scores. Predictive precision was assessed by correlating predicted and observed scores, with the significance of prediction performance evaluated through a permutation test.

RESULTS

The negative model of the CPM procedure demonstrated a significant capacity to predict MoCA scores (r = -0.309, p = 0.002). Similarly, the CPM procedure could predict MMSE scores (r = -0.236, p = 0.016). The predictive models for cognitive test scores in LLD primarily involved the visual network, somatomotor network, dorsal attention network, and ventral attention network.

CONCLUSIONS

Brain functional connectivity emerges as a promising predictor of personalized cognitive test scores in LLD, suggesting that functional connectomes are potential neurobiological markers for cognitive performance in patients with LLD.

NLRP3-GABA signaling pathway contributes to the pathogenesis of impulsive-like behaviors and cognitive deficits in aged mice

BACKGROUND

Perioperative neurocognitive disorders (PND), such as delirium and cognitive impairment, are commonly encountered complications in aged patients. The inhibitory neurotransmitter γ-aminobutyric acid (GABA) is aberrantly synthesized from reactive astrocytes following inflammatory stimulation and is implicated in the pathophysiology of neurodegenerative diseases. Additionally, the activation of NOD-like receptor protein 3 (NLRP3) inflammasome is involved in PND. Herein, we aimed to investigate whether the NLRP3-GABA signaling pathway contributes to the pathogenesis of aging mice's PND.

METHODS

24-month-old C57BL/6 and astrocyte-specific NLRP3 knockout male mice were used to establish a PND model via tibial fracture surgery. The monoamine oxidase-B (MAOB) inhibitor selegiline (1 mg/kg) was intraperitoneally administered once a day for 7 days after the surgery. PND, including impulsive-like behaviors and cognitive impairment, was evaluated by open field test, elevated plus maze, and fear conditioning. Thereafter, pathological changes of neurodegeneration were explored by western blot and immunofluorescence assays.

RESULTS

Selegiline administration significantly ameliorated TF-induced impulsive-like behaviors and reduced excessive GABA production in reactive hippocampal astrocytes. Moreover, astrocyte-specific NLRP3 knockout mice reversed TF-induced impulsive-like and cognitive impairment behaviors, decreased GABA levels in reactive astrocytes, ameliorated NLRP3-associated inflammatory responses during the early stage, and restored neuronal degeneration in the hippocampus.

CONCLUSIONS

Our findings suggest that anesthesia and surgical procedures trigger neuroinflammation and cognitive deficits, which may be due to NLRP3-GABA activation in the hippocampus of aged mice.

Predicting sex, age, general cognition and mental health with machine learning on brain structural connectomes

There is an increasing expectation that advanced, computationally expensive machine learning (ML) techniques, when applied to large population-wide neuroimaging datasets, will help to uncover key differences in the human brain in health and disease. We take a comprehensive approach to explore how multiple aspects of brain structural connectivity can predict sex, age, general cognitive function and general psychopathology, testing different ML algorithms from deep learning (DL) model (BrainNetCNN) to classical ML methods. We modelled N = 8183 structural connectomes from UK Biobank using six different structural network weightings obtained from diffusion MRI. Streamline count generally provided the highest prediction accuracies in all prediction tasks. DL did not improve on prediction accuracies from simpler linear models. Further, high correlations between gradient attribution coefficients from DL and model coefficients from linear models suggested the models ranked the importance of features in similar ways, which indirectly suggested the similarity in models' strategies for making predictive decision to some extent. This highlights that model complexity is unlikely to improve detection of associations between structural connectomes and complex phenotypes with the current sample size.

Connectome-based predictive modeling predicts paranoid ideation in young men with paranoid personality disorder: a resting-state functional magnetic resonance imaging study

Paranoid personality disorder (PPD), a mental disorder that affects interpersonal relationships and work, is frequently neglected during diagnosis and evaluation at the individual-level. This preliminary study aimed to investigate whether connectome-based predictive modeling (CPM) can predict paranoia scores of young men with PPD using whole-brain resting-state functional connectivity (rs-FC). College students with paranoid tendencies were screened using paranoia scores ≥60 derived from the Minnesota Multiphasic Personality Inventory; 18 participants were ultimately diagnosed with PPD according to the Diagnostic and Statistical Manual of Mental Disorders and subsequently underwent resting-state functional magnetic resonance imaging. Whole-brain rs-FC was constructed, and the ability of this rs-FC to predict paranoia scores was evaluated using CPM. The significance of the models was assessed using permutation tests. The model constructed based on the negative prediction network involving the limbic system-temporal lobe was observed to have significant predictive ability for paranoia scores, whereas the model constructed using the positive and combined prediction network had no significant predictive ability. In conclusion, using CPM, whole-brain rs-FC predicted the paranoia score of patients with PPD. The limbic system-temporal lobe FC pattern is expected to become an important neurological marker for evaluating paranoid ideation.