This repository is part of my Final Year Project, titled: "Deep Learning-Based Image Segmentation of Membranous Urethra in MR Imaging"
It contains the following folders:
/data
- contains the slice images and binary segmentation masks in different formats (both lines, right lines, different thicknesses, closed area), as well as Patient ID and MUL value information ininfo.csv
/data_processing
- data preprocessing for initial DICOM images to NumPy/model
- ML pipeline for model training and evaluation, for both segmentation and regression