CUHK-AIM-Group
Official@cuhk-aim-group · Hong Kong
The Artificial Intelligence in Medicine (AIM) Group from The Chinese University of Hong Kong (CUHK)
Agent Skills by CUHK-AIM-Group
Showing 102 vetted skills indexed across 2 GitHub repositories.
CUHK-AIM-Groupcobre-skill
Orchestrates COBRE dataset download, BIDS organization, and sMRI/rs-fMRI processing for schizophrenia research.
CUHK-AIM-Groupneuroimaging-decoding
Runs ROI MVPA, ROI-wise GLM, and voxel-wise SearchLight decoding on neuroimaging data.
CUHK-AIM-Grouphierarchical
Generates data-driven brain parcellations from neuroimaging features using hierarchical clustering.
CUHK-AIM-Groupibgnn
Trains IBGNN graph neural networks on fMRI connectome data for phenotype prediction.
CUHK-AIM-Groupneurostorm
Preprocess, pretrain, fine-tune, and run inference on fMRI data with eight deep learning models.
CUHK-AIM-Groupbrain-age-modeling
Trains cross-validated brain-age regression models and exports bias-corrected Brain-PAD predictions.
CUHK-AIM-Groupfm_app
Runs the FM-APP multi-stage pipeline for phenotype prediction from fMRI and sMRI ROI features.
CUHK-AIM-Groupbrainnetcnn
Train BrainNetCNN on dense ROI connectivity matrices for neuroimaging classification and regression.
CUHK-AIM-Groupnsd-skill
Orchestrates BIDS validation, multimodal processing, and stimulus extraction for the Natural Scenes Dataset.
CUHK-AIM-Groupabcd-skill
Orchestrates ABCD Study data download, BIDS staging, and multimodal MRI processing workflows.
CUHK-AIM-Grouplggnn
Trains LG-GNN graph neural networks on fMRI brain graphs for phenotype prediction.
CUHK-AIM-Grouprest-mneta-mdd-skill
Orchestrates BIDS validation, rs-fMRI processing, phenotype extraction, and QC for the REST-meta-MDD dataset.
CUHK-AIM-Grouphcpya-skill
Orchestrates HCP Young Adult dataset download, BIDS staging, and multimodal MRI processing workflows.
CUHK-AIM-Groupcnn3d
Trains a compact residual 3D CNN for voxel-level classification and regression on volumetric neuroimaging data.
CUHK-AIM-Grouphcpa-skill
Orchestrates HCP Aging dataset download, BIDS staging, and multimodal MRI processing workflows.
CUHK-AIM-Groupaibl-skill
Orchestrates AIBL dataset download, BIDS staging, and multimodal MRI and PET processing workflows.
CUHK-AIM-Groupmschallenge-skill
Orchestrates validation, lesion analysis, and QC for the Longitudinal MS Lesion Segmentation Challenge dataset.
CUHK-AIM-Groupseed-iv-skill
Orchestrates SEED-IV EEG validation, feature extraction, and emotion classification workflows.
CUHK-AIM-Groupmeg-skill
Processes MEG data for preprocessing, time-frequency analysis, source localization, and connectivity analysis.
CUHK-AIM-Groupfiltering
Applies temporal high-pass, low-pass, and band-pass filtering to denoise preprocessed fMRI BOLD time series.
CUHK-AIM-Grouphcpd-skill
Orchestrates HCP Development dataset download, BIDS staging, and multimodal MRI processing workflows.
CUHK-AIM-Groupbraingnn
Train and evaluate BrainGNN graph neural networks on fMRI connectivity matrices for phenotype prediction.
CUHK-AIM-Groupbnt
Trains BrainNetworkTransformer on dense fMRI functional connectivity matrices for phenotype prediction.
CUHK-AIM-Grouptemporal-models
Trains LSTM, GRU, TCN, and Transformer encoders on longitudinal neuroimaging sequences.
Frequently Asked Questions About CUHK-AIM-Group
FAQPage SchemaWhat specific neuroimaging tasks are supported by these capabilities?▼
These capabilities support DICOM to NIfTI conversion, structural MRI segmentation, EEG preprocessing, and fMRI connectivity analysis. They enable standardized BIDS-compliant data organization and the execution of complex neuroimaging pipelines including FreeSurfer, FSL, and QSIPrep.
Who is the target persona for these neuroimaging resources?▼
The target personas are clinical researchers, neuroscientists, and medical imaging engineers. These resources are designed for professionals managing large-scale neuroimaging datasets who require reproducible, audit-logged, and structured processing environments for clinical studies.
What are the prerequisites for running these neuroimaging pipelines?▼
Users require a Unix-based environment with Docker and Conda installed to manage containerized dependencies. The system relies on BIDS-formatted datasets and specific neuroimaging software suites like FSL, FreeSurfer, or CONN, which must be pre-installed or accessible via the provided container management modules.