Neuroimaging Workflow for Parkinson's Disease Subjects

This repository provides a MATLAB-based workflow (WorkFlow_newPDSubject.m) for the analysis and quantification of neuroimaging biomarkers in Parkinson's Disease (PD) subjects. The pipeline integrates widely-used neuroimaging tools, including SPM (Statistical Parametric Mapping) and Fiji, to process raw DICOM data through to quantitative brain structure analysis.

---

Workflow Overview

The main script, WorkFlow_newPDSubject.m, executes the full processing pipeline for PD subjects in sequential steps:

1. Load and Quality Check New PD Subject Protocol Sequences

• Detects new subjects from the raw DICOM database.
• Identifies sequences of interest (e.g., 3D, SWI).
• Performs an initial acquisition quality check.

2. Preprocessing with SPM Toolbox

• Converts DICOM files to NIfTI format.
• Performs realignment and reslicing of images.
• Corrects bounding boxes and verifies registration with a reference atlas subject.
• Applies transformation matrices for coregistration.

3. Fiji Preprocessing

• Converts images to 32-bit format.
• Performs vertical flipping and converts NIfTI files to Analyze HDR format.

4. Obtain Atlas Results from Cluster (3D and SWI Sequences)

• Applies label fusion methods (Majority Voting, Weighted Voting, Local Weighted Voting) using a reference atlas to segment key brain structures.

5. Multimodal Registration (SWI → 3D)

• Registers Susceptibility Weighted Imaging (SWI) sequences to 3D sequences using rigid multimodal registration.

6. Extract Brainstem Segmentation Metrics

• Calculates and extracts segmentation metrics for brainstem structures using Fiji.

---

Dependencies

The workflow requires the following software:

• MATLAB – Main programming environment.
• SPM (Statistical Parametric Mapping) – Brain imaging analysis.
• Fiji (ImageJ) – Image processing for preprocessing and segmentation steps.
• utils/ – Custom MATLAB utility functions required for the script.

---

Usage

1. Clone this repository to your local machine:

   git clone https://github.com/CIMA-Imaging-Platform/DLB_Image_Biomarkers.git
   

Ensure MATLAB, SPM, and Fiji are installed and configured.

Update paths in WorkFlow_newPDSubject.m to match your local directory structure. Key paths include:

utils

resultsCluster

dicomDir

Run the script in MATLAB:

WorkFlow_newPDSubject

License This project is licensed under GNU GENERAL PUBLIC LICENSE Version 3, 29 June 2007 file.