Out-of-body image detection in surgical videos

Research Group CAMMA / University of Strasbourg / IHU Strasbourg

http://camma.u-strasbg.fr/

Out-of-body frames in endoscopic surgeries can contain privacy sensitive information. This tool is meant to help protect privacy by detecting and blurring out these out-of-body frames. Its performance is reported in the publication mentioned below. This tool is provided for demonstration and without warranty. The authors or their institutions can not be held liable for any privacy concern due to undetected out-of-body frames. This tool can be used through either the command-line interface or the GUI application. If you're using Windows 10, you can download and run the executable directly from here.

Usage

From command-line

Execute the OOBNet by passing the input and output video paths in the command-line

python oobnet_exec.py --video_in <input/video/path> --video_out <output/video/path>

Optionally, you can also save the results to a text file. The file will contain frame ids and raw prediction results.

python oobnet_exec.py --video_in <input/video/path> --video_out <output/video/path> --text_out <output/text/path>

GUI

Alternatively, you can launch the GUI application by running the following command:

python oobnet_gui.py

Installation

1. Install Anaconda on your computer if you don't already have it. You can download it from here.

2. Clone this repository then cd to its directory on your computer. Download the model checkpoint

wget -P ckpt https://s3.unistra.fr/camma_public/github/oobnet_detection/ckpt/oobnet_weights.h5

3. Create a new conda environment conda create --name oob_detection python==3.8.5

4. Activate the environment conda activate oob_detection

5. Install dependencies pip3 install -r requirements.txt

Note: If you have a GPU, you can replace tensorflow with tensorflow-gpu in the requirements.txt file before the 5th step.

Credits

When referring to this software, please cite the following publication:

@article{lavanchy_preserving_2023,
	title = {Preserving privacy in surgical video analysis using a deep learning classifier to identify out-of-body scenes in endoscopic videos},
	volume = {13},
	issn = {2045-2322},
	url = {https://doi.org/10.1038/s41598-023-36453-1},
	doi = {10.1038/s41598-023-36453-1},
	number = {1},
	journal = {Scientific Reports},
	author = {Lavanchy, Joël L. and Vardazaryan, Armine and Mascagni, Pietro and AI4SafeChole Consortium and Mutter, Didier and Padoy, Nicolas},
	month = jun,
	year = {2023},
	pages = {9235},
}

License

This code is available for non-commercial scientific research purposes as defined in the CC BY-NC-SA 4.0. By downloading and using this code you agree to the terms in the LICENSE. Third-party codes are subject to their respective licenses.

This license allows reusers to distribute, remix, adapt, and build upon the material in any medium or format for noncommercial purposes only, and only so long as attribution is given to the creator. If you remix, adapt or build upon the material, you must license the modified material under identical terms.