niitools 0.0.4

My humble package for NeuroImaging Informatics tools

niitools

NeuroImaging Informatics Tools — Extract and aggregate brain segmentation statistics from FastSurfer outputs

Overview

niitools is a Python package designed to streamline the extraction and aggregation of neuroimaging statistics from FastSurfer brain segmentation outputs. It processes multiple subjects at scale and generates publication-ready CSV and Excel reports for four key neuroimaging metrics:

• Cortical Thickness — Average thickness across cortical regions (left & right hemispheres)
• Subcortical Segmentation — Volumes of subcortical structures (Aseg)
• Brain Stem Analysis — Volumetric measurements of brain stem components
• Cerebellar Segmentation — Cerebellar structure volumes via CerebNet

Perfect for neuroimaging research pipelines, clinical studies, and batch processing of brain MRI analysis outputs.

---

✨ Features

• 🚀 Batch Processing — Process hundreds of subjects in a single command
• 📊 Multi-Format Output — Generate both CSV and Excel files with automatic formatting
• 🧠 Four Analysis Modules — Comprehensive coverage of cortical, subcortical, brainstem, and cerebellar metrics
• 🔄 Flexible API — Use as a command-line tool or import as a Python library
• 📝 Logging Support — Detailed logging with loguru for debugging and audit trails
• ✅ Validation — Automatic validation of FastSurfer output directory structure

---

📦 Installation

From Source (Recommended for Development)

git clone https://github.com/diogohs/niitools.git
cd niitools
pip install -e .

This installs the package in editable mode and registers the niitools command-line tool.

From PyPI

pip install niitools

Requirements

• Python: 3.8 or higher
• Dependencies:
  • pandas>=2.0.3 — Data manipulation and analysis
  • openpyxl>=3.1.5 — Excel file generation
  • loguru>=0.7.3 — Advanced logging

---

🚀 Quick Start

Command-Line Usage

Extract FastSurfer statistics from all subjects in a directory:

niitools fastsurfer -i ./data/input -o ./data/output

Arguments:

• -i, --input (required) — Path to directory containing FastSurfer output directories
• -o, --output (required) — Path to output directory for CSV and Excel files

Example:

niitools fastsurfer -i /mri_data/fastsurfer_results -o /mri_data/statistics

Python Library Usage

Import and use niitools in your Python scripts:

from pathlib import Path
from niitools.fastsurfer import run_fastsurfer

# Define input and output paths
input_dir = Path("./data/input")
output_dir = Path("./data/output")

# Run the analysis
run_fastsurfer(input_dir, output_dir)

The function will:

1. Scan the input directory for valid FastSurfer output structures
2. Extract statistics from each subject
3. Aggregate results across all subjects
4. Generate both CSV and Excel output files

---

📋 Output Files

The analysis generates 8 output files (4 metrics × 2 formats):

Files Generated

Metric	CSV File	Excel File
Cortical Thickness (Left)	FastSurfer_Cortical_Thickness_left.csv	FastSurfer_Cortical_Thickness_left.xlsx
Cortical Thickness (Right)	FastSurfer_Cortical_Thickness_right.csv	FastSurfer_Cortical_Thickness_right.xlsx
Subcortical (Aseg)	FastSurfer_Aseg.csv	FastSurfer_Aseg.xlsx
Brain Stem	FastSurfer_BrainStem.csv	FastSurfer_BrainStem.xlsx
Cerebellar (CerebNet)	FastSurfer_CerebNet.csv	FastSurfer_CerebNet.xlsx

Output Structure

Each output file contains:

• Subject Column — Subject identifier in the first column
• Region Columns — One column per brain region with extracted metrics
• One Row per Subject — Aggregated statistics for each subject

---

🏗️ Project Structure

niitools/
├── niitools/
│   ├── __init__.py              # Package initialization
│   ├── __main__.py              # CLI entry point
│   ├── validate.py              # Validation utilities
│   ├── utils.py                 # Helper functions
│   └── fastsurfer/
│       ├── __init__.py          # Module exports
│       ├── _executor.py         # Main execution logic
│       ├── aseg.py              # Subcortical segmentation parsing
│       ├── brainstem.py         # Brain stem parsing
│       ├── cerebnet.py          # Cerebellar segmentation parsing
│       └── cortical.py          # Cortical thickness parsing
├── tests/
│   ├── __init__.py
│   └── test_converter.py        # Unit tests
├── notebooks/                   # Jupyter notebooks with examples
├── data/
│   ├── input/                   # FastSurfer outputs (input)
│   └── output/                  # Generated statistics (output)
├── pyproject.toml               # Project configuration
├── README.md                    # This file
└── LICENSE                      # MIT License

---

📚 Advanced Usage

Processing with Custom Logging

from pathlib import Path
from loguru import logger
from niitools.fastsurfer import run_fastsurfer

# Configure logging
logger.add("niitools_processing.log", level="DEBUG")

# Run with logging enabled
run_fastsurfer(Path("./input"), Path("./output"))

Validating FastSurfer Directories

from pathlib import Path
from niitools.validate import is_fastsurfer_output

subject_dir = Path("./data/SUBJECT_001")
if is_fastsurfer_output(subject_dir):
    print("Valid FastSurfer output structure!")
else:
    print("Missing required subdirectories")

---

⚙️ FastSurfer Output Requirements

niitools expects standard FastSurfer output directories with the following structure:

SUBJECT_ID/
├── label/          # Brain region labels
├── mri/            # MRI volumes and brain stem outputs
├── scripts/        # FastSurfer processing scripts
├── stats/          # Statistical output files
└── surf/           # Surface files

Ensure your FastSurfer output contains these required files:

• stats/aseg.stats — Subcortical segmentation statistics
• stats/cerebellum.CerebNet.stats — Cerebellar segmentation
• stats/lh.aparc.DKTatlas.mapped.stats — Left hemisphere cortical thickness
• stats/rh.aparc.DKTatlas.mapped.stats — Right hemisphere cortical thickness
• mri/brainstemSsLabels.volumes.txt — Brain stem volumes

---

🐛 Troubleshooting

No subjects found

⚠️  No valid FastSurfer output directories found in {input_dir}

Solution: Verify that input directories contain the required FastSurfer subdirectories (label, mri, scripts, stats, surf).

Missing statistics files

⚠️  Missing required statistics file

Solution: Ensure FastSurfer processing completed successfully and generated all required output files.

Python version incompatibility

ERROR: This package requires Python 3.8 or higher

Solution: Upgrade Python: python --version should return 3.8+

---

🤝 Contributing

Contributions are welcome! Please feel free to:

• Report bugs via GitHub Issues
• Submit pull requests for improvements
• Suggest new features or analysis modules

---

📄 License

This project is licensed under the MIT License — see the LICENSE file for details.