A tool for the Automated Extraction of Pennation angles in Ultrasound images with low Signal-to-noise ratio
This repository contains the work in progress on the AEPUS tool.
The current implementation performs physiological feature extraction from Ultrasound (US) images of gastrocnemius muscle.
AEPUS identifies:
- Superficial Aponeurosis (blue)
- Deep Aponeurosis (red)
- Average Fascicle inclination angle (green)
This repository contains:
-
aepusfolder contains the source code with low-level routines. -
testsfolder contains- a simple test demonstrating feature extraction
- sample US images required for the test (located in
datadirectory)
To install the library we advise using miniconda package manager and creating a virtual environment. To do it:
- Download the repository to your local PC
- Install miniconda
- Move to the library directory
- Execute the command below to create a virtual environment named
pybf_envto install all necessary libraries listed inconda_requirements.txt
conda create --name aepus_env python=3.9 --file conda_requirements.txtNote: If you have problems with installing the packages automatically you can do it manually. Critical packages for the library are:
- numpy
- matplotlib
- scipy
- scikit-image
- pillow
- opencv (for visualization)
To use existing features we advise exploring the provided test.
To run the test:
- Run a terminal and activate conda environment
conda activate aepus_env- Navigate to the directory of the test
- Execute
python main.py
All source code is released under Apache v2.0 license unless noted otherwise, please refer to the LICENSE file for details.
Example datasets under tests/data folder are provided under a Creative Commons Attribution No Derivatives 4.0 International License