PVSBenchmark

This repository contains a Docker image that runs a benchmark for the Frangi filter applied to medical images. The image processes NIfTI files and provides multiple thresholds to evaluate performance.

Usage

Follow the steps below to pull the Docker image and run the benchmark and Frangi filter.

Step 1: Pull the Docker Image

First, download the Docker image from the repository by running the following command:

docker pull camijaramillo/benchmark_pvs:Thresholds

Step 2: Run the Docker Container

Next, start the Docker container, mounting your local folder to share files between your computer and Docker. Use this command:

docker run --rm -it -v "filelocation:/mnt/data" camijaramillo/benchmark_pvs:Thresholds /bin/bash

Explanation:

• --rm: Automatically removes the container after it stops.
• -it: Opens an interactive terminal.
• -v "filelocation:/mnt/data": Mounts your local directory (C:\Users\filelocation) to the /mnt/data directory inside the container, enabling access to files.

Step 3: Navigate to the Build Directory

Once inside the container, navigate to the build directory where the benchmark executable is located:

cd build

Step 4: Run the Benchmark

Run the benchmark using a list of images and parameters stored in text files (imagenes.txt and params.txt), along with the number of thresholds to apply:

./itk_frangi_benchmark /mnt/data/imagenes.txt /mnt/data/params.txt Frangi 10

Explanation:

• /mnt/data/imagenes.txt: Path to the text file containing image file paths.
• /mnt/data/params.txt: Path to the text file containing the parameters for the benchmark.
• Frangi: Specifies the Frangi filter as the method to use.
• 10: The number of thresholds to apply.

Step 5: Run the Frangi Filter

To apply the Frangi filter to a specific image, use the following command:

./itk_frangi_filter --input /mnt/data/AAA417_56_hm_tse_ac-pc_512_p2_reg2ref_Warped.nii --output /mnt/data/salida2.nii --sigmaMin 0.2 --sigmaMax 0.4 --nbSigmaSteps 4 --alpha 3.0 --beta 0.1 --gamma 1000

Explanation:

• input: Path to the input NIfTI image (.nii format).
• output: Path to the output image where the result will be saved.
• sigmaMin and --sigmaMax: Define the range of scales (sigma) to apply to the filter.
• nbSigmaSteps: Number of steps between sigmaMin and sigmaMax.
• alpha, --beta, and --gamma: Parameters that adjust the filter’s response to tubular structures.