Biology subjects

Chen, N.-X.

Publications and source records attributed to Chen, N.-X..

4 recordsLinked to original sources

bioRxiv2020

The Contribution of Brain Structural and Functional Variance in Predicting Age, Sex and Treatment

Structural and functional neuroimaging have been widely used to track and predict demographic and clinical variables, including treatment outcomes. However, it is often difficult to directly establish and compare the respective weights and contributions of brain structure and function in prediction studies. The present study aimed to directly investigate respective roles of brain structural and functional indices, along with their contributions in the prediction of demographic variables (age/sex) and clinical changes of schizophrenia patients. The present study enrolled 492 healthy people from Southwest University Adult Lifespan Dataset (SALD) for demographic variables analysis and 42 patients with schizophrenia from West China Hospital for treatment analysis. We conducted a model fit test with two variables (one voxel-based structural metric and another voxel-based functional metric) and then performed a variance partitioning on the voxels that can be predicted sufficiently. Permutation tests were applied to compare the contribution difference between each pair of structural and functional measurements. We found that voxel-based structural indices had stronger predictive value for age and sex, while voxel-based functional metrics showed stronger predictive value for treatment. Therefore, through variance partitioning, we could clearly and directly explore and compare the voxel-based structural and functional indices on particular variables. In sum, for long-term change variable (age) and constant biological feature (sex), the voxel-based structural metrics would contribute more than voxel-based functional metrics; but for short-term change variable (schizophrenia treatment), the functional metrics could contribute more.

neuroscience↗

bioRxiv2020

An Industrial-Grade Brain Imaging-Based Deep Learning Classifier

Beyond detecting brain lesions or tumors, comparatively little success has been attained in identifying brain disorders such as Alzheimers disease (AD), based on magnetic resonance imaging (MRI). Many machine learning algorithms to detect AD have been trained using limited training data, meaning they often generalize poorly when applied to scans from previously unseen populations. Therefore, we built a practical brain MRI-based AD diagnostic classifier using deep learning/transfer learning on a dataset of unprecedented size and diversity. A retrospective MRI dataset pooled from more than 217 sites/scanners constituted one of the largest brain MRI samples to date (85,721 scans from 50,876 participants) between January 2017 and August 2021. Next, a state-of-the-art deep convolutional neural network, Inception-ResNet-V2, was built as a sex classifier with high generalization capability. The sex classifier achieved 94.9% accuracy and served as a base model in transfer learning for the objective diagnosis of AD. After transfer learning, the model fine-tuned for AD classification achieved 90.9% accuracy in leave-sites-out cross-validation on the Alzheimers Disease Neuroimaging Initiative (ADNI, 6,857 samples) dataset and 94.5%/93.6%/91.1% accuracy for direct tests on three unseen independent datasets (AIBL, 669 samples / MIRIAD, 644 samples / OASIS, 1,123 samples). When this AD classifier was tested on brain images from unseen mild cognitive impairment (MCI) patients, MCI patients who converted to AD were 3 times more likely to be predicted as AD than MCI patients who did not convert (65.2% vs 20.6%). Predicted scores from the AD classifier showed significant correlations with illness severity. In sum, the proposed AD classifier offers a medical-grade marker that has potential to be integrated into AD diagnostic practice.

neuroscience↗

bioRxiv2019

Reproducible observations revealing patterns of the default mode network underlie rumination

Rumination is a repetitive self-referential thinking style and posited to be an expression of abnormalities in the default mode network (DMN) in major depressive disorder (MDD). Recent evidences indicate DMN is not a unitary network but can be further divided into 3 functionally heterogenous subsystems. However, the subsystem mechanism through which DMN underlie rumination remain unclear. Here, with a modified continuous state-like paradigm, we induced healthy participants to ruminate or imagine objective scenarios (as a distraction control condition) on 3 different MRI scanners. We compared functional connectivities (FC) and inter-subject correlations (ISC) of the DMN and its 3 subsystems between rumination and distraction states. Results yielded a highly reproducible and dissociated pattern. During rumination, within-DMN FC was generally decreased compared to the distraction state. At the subsystem level, we found increased FC between the core and medial temporal lobe (MTL) subsystem and decreased FC between the core and dorsal medial prefrontal cortex (DMPFC) subsystem and within the MTL subsystem. Furthermore, we found decreased ISC within the MTL subsystem. These results suggest a specific and less synchronous activity pattern of DMN during rumination and shed new light on the association between rumination and DMN substrates regarding MDD.

neuroscience↗

bioRxiv2018

Reduced but not Enhanced Default Mode Network Functional Connectivity in Major Depressive Disorder: Evidence from 25 Cohorts in the REST-meta-MDD Project

Major Depressive Disorder (MDD) is common and disabling, but its neural pathophysiology remains unclear. Functional brain network studies in MDD have largely had limited statistical power and data analysis approaches have varied widely. The REST-meta-MDD Project of resting-state fMRI (R-fMRI) addresses these issues. The 25 research groups in China composing the REST-meta-MDD Project contributed R-fMRI data of 1,300 patients with MDD and 1,128 normal controls (NCs). The data were preprocessed locally with a standardized protocol prior to aggregated group analyses. We focused on functional connectivity (FC) within the default mode network (DMN), frequently reported to show increased FC in MDD. We found decreased instead of increased DMN FC when comparing 848 MDDs with 794 NCs from 17 sites after data exclusion. We found FC reduction only in recurrent MDD, not in first-episode drug-naive MDD. Decreased DMN FC was associated with medication usage but not with MDD duration. DMN FC was also positively related to symptom severity but only in recurrent MDDs. Exploratory analyses also revealed alterations of local intrinsic activity in MDD. We confirmed the key role of DMN in MDD but found reduced rather than increased FC within the DMN. Future studies should test whether decreased DMN FC mediates treatment response. This manuscript announces the publicly available resting-state fMRI indices of the REST-meta-MDD consortium shared via the R-fMRI Maps Project.\n\nSIGNIFICANCE STATEMENTFunctional connectivity within the default mode network in major depressive disorder patients has been frequently reported abnormal but with contradicting directions in previous small sample size studies. By creating the REST-meta-MDD consortium containing neuroimaging data of 1,300 depressed patients and 1,128 normal controls from 25 research groups in China, we found decreased default mode network functional connectivity in depressed patients, driven by patients with recurrent depression, and associated with current medication treatment but not with disease duration. These findings suggest that default mode network functional connectivity remains a prime target for understanding the pathophysiology of depression, with particular relevance to revealing mechanisms of effective treatments.

neuroscience↗