Skip to content

Latest commit

 

History

History
25 lines (20 loc) · 1.29 KB

README.md

File metadata and controls

25 lines (20 loc) · 1.29 KB

skull_stripping_project

The aim of the project is to prepare a skull stripping method on T1 magnetic resonance imaging (MRI) images of the head. The method should extract the entire area occupied by the brain proper from the patient's raw scan, excluding bones, other soft tissues, etc.

I divided the project into three parts, each of them is contained in a separate one Google Colab:

  • create_dataset -> from downloading data from the source to saving the finished dataset to training the neural network
  • train_model -> train model and save it
  • model_predict -> generate predictions for test data (at the moment I do not have access to ground true of test data so I cannot define the percentage of network efficiency)

Training

Results

Test on data from a different source:

(:point_up: It's me :grinning:)