https://github.com/pegah-ardehkhani/brain-mri-segmentation

U-Net from Scratch for Brain Tumor Segmentation
https://github.com/pegah-ardehkhani/brain-mri-segmentation

ai-for-medical-diagnosis brain-imaging brain-mri brain-mri-images deep-learning deep-neural-networks dice-coefficient healthcare-imaging iou medical-image-processing medical-imaging mri mri-brain mri-brain-segmentation mri-images python segmentation tensorflow u-net u-net-keras

Last synced: 2 months ago
JSON representation

U-Net from Scratch for Brain Tumor Segmentation

Host: GitHub
URL: https://github.com/pegah-ardehkhani/brain-mri-segmentation
Owner: Pegah-Ardehkhani
License: mit
Created: 2022-07-11T18:54:07.000Z (almost 3 years ago)
Default Branch: main
Last Pushed: 2022-07-11T19:57:05.000Z (almost 3 years ago)
Last Synced: 2025-01-07T17:18:32.966Z (4 months ago)
Topics: ai-for-medical-diagnosis, brain-imaging, brain-mri, brain-mri-images, deep-learning, deep-neural-networks, dice-coefficient, healthcare-imaging, iou, medical-image-processing, medical-imaging, mri, mri-brain, mri-brain-segmentation, mri-images, python, segmentation, tensorflow, u-net, u-net-keras
Language: Jupyter Notebook
Homepage:
Size: 8 MB
Stars: 3
Watchers: 1
Forks: 0
Open Issues: 0
Metadata Files:
- Readme: README.md
- License: LICENSE

Awesome Lists containing this project

README

# Brain MRI Segmentation 🧠 ![license](https://img.shields.io/github/license/Pegah-Ardehkhani/Brain-MRI-Segmentation.svg) ![releases](https://img.shields.io/github/release/Pegah-Ardehkhani/Brain-MRI-Segmentation.svg) [![nbviewer](https://img.shields.io/badge/render-nbviewer-orange.svg)](http://nbviewer.org/github/Pegah-Ardehkhani/Brain-MRI-Segmentation/blob/main/Brain%20MRI%20Segmentation.ipynb)

## Dataset 📔

[Kaggle link: Brain Tumor Data](https://www.kaggle.com/mateuszbuda/lgg-mri-segmentation)

The dataset used for development was obtained from The Cancer Imaging Archive (TCIA) and involved 110 cases of lower-grade glioma patients. Registers brain MR images with manual FLAIR abnormality segmentation masks are published as a Kaggle Dataset lgg-mri-segmentation.

## Project Overview

**U-net:**

U-net from scratch has been written.

Key aspects of U-Net:

1. Convolution Layers: Convolution operation are used to learn information from images which then can be used as features for machine learning problems.

2. Down Sampling: Sequence of convolution combined with max pooling results in down sampling. In down sampling, size of the image is reduced which means we can observe larger portion of image in a single convolution operation. Down sampling is a good approach for identifying what is present in the image but for identifying where the object is we need to use upsampling.

3. Up Sampling: It is just opposite of down sampling. We go from low resolution to high resolution. For up sampling U-Net uses transposed covolution which is achieved by taking transpose of filter kernels and reversing the process of convolution.

Following picture gives a clear picture of What a U-Net is.