oasis-skill

Orchestrates BIDS validation, sMRI processing, and phenotype extraction for OASIS neuroimaging datasets.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve? Processing the OASIS (Open Access Series of Imaging Studies) dataset for aging and Alzheimer's disease research requires coordinating BIDS validation, structural MRI preprocessing, and clinical phenotype extraction across multiple tools, which is error-prone when done manually. ## Core Features & Use Cases - BIDS Validation: Validates OASIS-1, OASIS-2, and OASIS-3 directory structures, detects dataset version, and reports modality completeness per subject. - Phenotype Extraction: Merges CDR, MMSE, demographic, and clinical data from multiple files into a single subject-aligned CSV table. - QC Summaries: Generates per-subject quality control reports with framewise displacement thresholds and exclusion lists, annotated with CDR/MMSE group info. - Use Case: A researcher downloads OASIS-3 data and needs to validate its BIDS structure, run cortical thickness analysis via sMRI processing, and merge CDR scores with imaging subject IDs before statistical analysis. ## Quick Start Ask the agent to validate the BIDS structure of your local OASIS dataset and extract the phenotype table with CDR and MMSE scores into a merged CSV.

Frequently Asked Questions about oasis-skill

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

FAQPage Schema
How do I validate OASIS BIDS data structure?

Run the validate_oasis.py script with --input pointing to your OASIS BIDS directory and --output for the report path. It detects the OASIS version, checks T1w completeness, and reports PET/ASL availability per subject in a CSV.

How to extract CDR and MMSE scores from OASIS phenotype files?

Use the extract_oasis_phenotype.py script with --phenotype-files listing your CSV/TSV phenotype files. It merges CDR, MMSE, age, sex, SES, and education columns by subject ID and can filter against an imaging subject list.

What is the difference between OASIS-1, OASIS-2, and OASIS-3?

OASIS-1 is cross-sectional with about 416 subjects aged 18-96 and T1w MRI. OASIS-2 is longitudinal with about 150 older subjects, and OASIS-3 adds over 1,000 subjects with longitudinal T1w, amyloid/tau PET, and ASL data.

Does this skill process PET or ASL data directly?

No, it is an orchestration-only skill that delegates modality processing to other skills. PET data goes to pet-skill for SUVR computation and ASL data goes to asl-skill for CBF quantification.

What are the limitations of OASIS-1 for neuroimaging analysis?

OASIS-1 was acquired on a 1.5T Siemens Vision scanner, so resolution is lower than modern 3T or 7T datasets. Its wide age range of 18-96 also means analyses should account for age effects.