aibl-skill

Orchestrates AIBL dataset download, BIDS staging, and multimodal MRI and PET processing workflows.

89|5|Updated Mar 25, 2026
One-click install
npx skills add https://github.com/CUHK-AIM-Group/NeuroDiscovery --skill aibl-skill-cuhk-aim-group
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: aibl-skill
Source: https://github.com/CUHK-AIM-Group/NeuroDiscovery/tree/main/skills/aibl-skill
Command: npx skills add https://github.com/CUHK-AIM-Group/NeuroDiscovery --skill aibl-skill-cuhk-aim-group

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pandas, and includes scripts (resource) components.

What problem does it solve? Working with the AIBL (Australian Imaging, Biomarkers and Lifestyle) dataset requires navigating portal-based data access, converting non-standard raw layouts into BIDS-compliant structure, merging scattered phenotype tables, and coordinating structural MRI and PET processing. This Skill provides a single orchestration layer that manages that entire workflow. ## Core Features & Use Cases - BIDS Reorganization: Converts raw AIBL subject/visit/modality folders into BIDS-compliant datasets with normalized subject IDs, session labels, PET tracer routing (PiB, FDG, tau), sidecar preservation, and dataset_description.json/participants.tsv generation. - Phenotype Extraction and Merging: Merges cognitive assessments, blood biomarkers, APOE genotype, and demographics into one analysis-ready CSV, with visit alignment, diagnosis normalization (HC/MCI/AD), and cross-referencing against imaging subject lists. - QC Summaries: Generates per-subject QC reports and exclusion lists from fMRIPrep confounds and FreeSurfer metrics with configurable motion thresholds. - Use Case: A researcher downloads raw AIBL data and asks to stage it into BIDS, merge phenotype tables for baseline visits, and produce a QC exclusion list before running statistical analysis on Alzheimer's disease biomarkers. ## Quick Start Ask the agent to reorganize your local raw AIBL directory into a BIDS-compliant dataset under aibl_output/bids and generate the participants and dataset metadata files.

Frequently Asked Questions about aibl-skill

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I convert AIBL raw data to BIDS format?

Run the reorganize_aibl.py script with --input pointing to your raw AIBL directory and --output for the BIDS target. It normalizes subject IDs like 002_S_0295 to sub-002S0295, maps visits to session labels, routes T1w and PET modalities, and supports a --dry-run preview mode.

How to merge AIBL phenotype tables for analysis?

Use extract_aibl_phenotype.py with --phenotype-dir pointing to your phenotype CSV/TSV files. It merges cognitive assessments, blood biomarkers, APOE genotype, and demographics on subject_id and visit, normalizes diagnosis labels to HC/MCI/AD, and can filter to subjects with imaging data.

Does AIBL data require special access to download?

Yes, AIBL imaging data requires registration and a data use agreement through the AIBL research portal, with downloads handled via the LONI Image Data Archive. The skill guides this access step but delegates actual download execution to shell tooling.

Can the BIDS reorganization handle both flat and nested AIBL layouts?

Yes, the script auto-detects whether input follows a flat subject/modality layout or a nested subject/visit/modality layout and processes each accordingly. PET files are routed with tracer labels for PiB, FDG, and tau based on filename patterns.

What QC metrics are used to exclude AIBL subjects?

The QC script collects framewise displacement from fMRIPrep confounds and completion status plus volume metrics from FreeSurfer outputs. Subjects are excluded when mean FD exceeds 0.3, max FD exceeds 5.0, or FreeSurfer recon-all did not complete, with thresholds configurable via arguments.