Accurate Liver Registration of 3D Ultrasound and CT Volume: An Open Dataset and a Model Fusion Method

This repository contains the code and dataset for the paper "Accurate Liver Registration of 3D Ultrasound and CT Volume: An Open Dataset and a Model Fusion Method".

---

Overview

Liver registration between ultrasound (US) and computed tomography (CT) is critical for various medical applications, such as surgical navigation and interventional guidance. This work introduces:

• A new open dataset: Featuring paired 3D ultrasound and CT volumes with accurate liver segmentation annotations.
• A model fusion-based registration method: Combining deep learning and traditional optimization approaches for robust and accurate results.

---

Dataset

The dataset includes:

1. 3D Ultrasound (US) Volumes
2. 3D Computed Tomography (CT) Volumes

Data Format

• US Data: Provided in NIfTI (.nii) format.
• CT Data: Provided in NIfTI (.nii) format.
• label: All data are registered in pairs, and you can transform them freely. The transformation matrix is ​​your label.

Access

Please request access to download the dataset.

---

Setup

Prerequisites

• Python >= 3.8
• PyTorch >= 1.10
• Numpy, Scipy, SimpleITK, and other dependencies listed in requirements.txt.

Installation

1. Clone the repository:

   git clone https://github.com/your-username/3D-US-CT-Liver-Registration.git
   cd 3D-US-CT-Liver-Registration