RESPECT_Data_Prep

🚀 Overview

RESPECT_Data_Prep is a Python-based toolkit for preparing MRI data for the RESPECT image registration pipeline.
It provides a set of scripts for:

• Converting MRI NIfTI files into slice-wise MR DICOM with consistent geometry
• Batch coregistration of multi-modal MRI series
• Visual inspection through image and segmentation overlays
• Generating basic segmentation quality metrics
• Cleaning and harmonizing dataset directory structures

The goal of this repository is to standardize MRI preprocessing, reduce manual errors, and provide reproducible preparation steps before downstream registration and analysis.

---

📁 Repository Structure

File	Description
Inspect_T1_T2_nii2Dicom.py	Converts T1/T2 NIfTI volumes into slice-wise MR DICOM with realistic metadata
batch_coregister_full_terminal_log.py	Runs batch coregistration and logs full terminal output
batch_coregister_inplace.py	Performs coregistration directly in existing folders
batch_coregister_print_only.py	Prints coregistration commands without executing them
overlay.py	Visual overlay of two MRI images
overlay_Seg_metrics.py	Overlay of anatomical images and segmentations + metrics
remove_leading_dot.py	Utility to fix file names starting with a dot
.gitignore	Git ignore rules for Python projects

---

🧠 Motivation

MRI preprocessing for registration pipelines often involves repetitive and error-prone manual steps.
This repository centralizes common preparation tasks such as:

• Format harmonization (NIfTI → DICOM)
• Batch execution of coregistration
• Visual quality control
• Dataset cleanup

This allows faster experimentation and more reproducible results across datasets.

---

📦 Requirements

System Requirements

• Python 3.9 or higher
• Linux / macOS (Windows via WSL also works)

Python Dependencies

The main dependencies are:

• nibabel
• pydicom
• numpy
• matplotlib
• scipy

Install them using:

pip install nibabel pydicom numpy matplotlib scipy