Pulmonary Embolism Dicom EDA & Segmentation Expo: [Link]
Full Preprocessing Tutorial: [Link]
Hounsfield Unit (HU) - a measure of radiodensity
Window Level
Resampling - pixel spacing => resampling the full dataset to a certain isotropic
LUNA16 Preprocessing [Link]
Augmentation
3D Mostafa: [Link]
Lung Segmentation using Marker-Controlled Watershed Transformation [Link]
2D & 3D Lung segmentation [Link]
2) Architecture:
3) Post-Processing:
Papers:
Methods:
4) Kernels:
DSB-2017 kernels [Link]
References:
Re-check:
Lung Nodule News: https://www.iaslc.org/iaslc-news
Lung New Papers: https://researcher.watson.ibm.com/researcher/view_group_subpage.php?id=9815
Data for Data Science Bowl 2017: https://academictorrents.com/details/015f31a94c600256868be155358dc114157507fc
Plot in 3d: https://www.kaggle.com/rodenluo/crop-save-and-view-nodules-in-3d/comments
LUNA16 notebooks: https://notebook.community/mas-dse-greina/neon/luna16/old_code/LUNA16_loader
DSB2017: Pre-processing:
First place: Full preprocessing tutorial by Guido Zuidhof
Second place: First pass through data with 3D ConvNet by Sentdex
Third place: Candidate generation and LUNA16 preprocessing by ArnavJain
LUNA16
https://github.com/Rakshith2597/Lung-nodule-detection-LUNA-16
https://github.com/svella9/Semi-Supervised-Learning-To-Improve-Lung-Cancer-Detection
https://www.kaggle.com/aadhavvignesh/lung-segmentation-by-marker-controlled-watershed
https://www.kaggle.com/kmader/dsb-lung-segmentation-algorithm
https://www.kaggle.com/arnavkj95/candidate-generation-and-luna16-preprocessing
https://www.kaggle.com/alibaba19/pulmonary-embolism-dicom-eda-segmentation-expo
https://www.kaggle.com/anoukstein/lung-segmentation-combined-approach
2 collections: [To Tony later]