3D Medical Imaging Pre-processing All-you-need

Pre-processing for 3D Medical Imaging

From the last year of my undergrad studies, I was very queries about Biomedical Imaging. But until the starting my master I don’t have the chance to go deep into medical imaging. Like most people at the beginning, I also suffered and was a bit confused about a few things. In this post, I will try to easily explain/show commonly used pre-processing in medical imaging especially with 3D Nifti.

In this post, we will be using Public Abdomen Dataset From Multi-Atlas Labeling Beyond the Cranial Vault — Workshop and Challenge Link: https://www.synapse.org/#!Synapse:syn3193805/wiki/217789

Most of these codes are adopted from the DLTK library, That’s a great resource, definitely consider have a look. Reference: https://github.com/DLTK/DLTK

We will cover:

1. Reading Nifti Data and plotting
2. Different Intensity Normalization Approaches
3. Resampling 3D CT data
4. Cropping and Padding CT data
5. Histogram equalization
6. Maximum Intensity Projection (MIP)

if You want know about MRI Histogram Matching, Histogram Equalization and Registration, You could have a look to my repoTo Learn about Segmentation
* https://github.com/fitushar/Brain-Tissue-Segmentation-Using-Deep-Learning-Pipeline-NeuroNet
* https://github.com/fitushar/Registration-as-Data-Augumentation-for-CT--DATA

To Learn about Segmentation

* **Brain Tissue Segmentation**, 3D : https://github.com/fitushar/Brain-Tissue-Segmentation-Using-Deep-Learning-Pipeline-NeuroNet
* **Chest-Abdomen-Pelvis (Segmentation)** 3D DenseVnet :https://github.com/fitushar/DenseVNet3D_Chest_Abdomen_Pelvis_Segmentation_tf2

* **3D-Unet** : https://github.com/fitushar/3DUnet_tensorflow2.0

Libraries need

* SimpleITK
* NumPy
* scipy
* skimage
* cv2
* DLTK

Reading Nifti Data and plotting

Intensity Normalization

Resampling

Crop or Padding

Github repo at https://github.com/fitushar/3D-Medical-Imaging-Preprocessing-All-you-need

Hope some of you will find it useful.