MIDRC CRP-2: Machine intelligence algorithms from multi-modal, multi-institutional COVID-19 data (https://www.midrc.org/midrc-collaborating-research-projects/project-one-crp2)
Development Team: S.Candemir (candemirsema@gmail.com)
Modality: Chest X-ray
Lung Lobe Segmentation model based on U-Net[1,2]. The model is trained with JSRT data [3] and the corresponding lung masks (SCR data) [4]. The training images are enhanced and re-sized to 256 x 256 before feeding to the network. The model is trained at The Ohio State University Wexner Medical Center, Department of Radiology [5], using Python, Tensorflow Keras API, and trained on an NVIDIA QuadroGV100 system with CUDA/CuDNNv9 dependecies.
Requirements: Python, Tensorflow Keras API, SimpleITK, OpenCV, Numpy, Scikit-image, Matplotlib
Predict
locate CT sequences in Data folder
>>python lung_segment.py
- U-Net: Convolutional Networks for Biomedical Image Segmentation (https://lmb.informatik.uni-freiburg.de/people/ronneber/u-net/)
- U-Net Architecture implementation (https://github.com/imlab-uiip/lung-segmentation-2d/tree/master/Demo)
- Training Data: JSRT database (http://db.jsrt.or.jp/eng.php)
- Training Data: SCR reference lung boundaries (https://www.isi.uu.nl/Research/Databases/SCR/)
- Laboratory for Augmented Intelligence in Imaging, The Ohio State University Wexner Medical Center,Department of Radiology (http://aii.osu.edu/)