You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
Alexander Wong 0e9b86a8f0 resnet50 example and baseline 7 months ago
apex @ 82dac9c941 preliminary resnet50 classification 7 months ago
data added data processing & viewing code, wip dataset class 7 months ago
runs/ResNet50 resnet50 example and baseline 7 months ago
.gitignore Initial commit 7 months ago
.gitmodules preliminary resnet50 classification 7 months ago
LICENSE Initial commit 7 months ago
README.md added data processing & viewing code, wip dataset class 7 months ago
main.py debugging rsna 7 months ago
process_and_view_dicom.ipynb preliminary resnet50 classification 7 months ago
pytorch_resnet50_example.ipynb resnet50 example and baseline 7 months ago
resnet50.pth resnet50 example and baseline 7 months ago
submission_resnet50.csv resnet50 example and baseline 7 months ago

README.md

RSNA Hemorrhage Classification

Source code for University of Alberta, CMPUT 617: Medical Image Analysis (Fall 2019). Lead by Prof. Nilanjan Ray.

Running Instructions

  1. Download and extract the RSNA intracranial hemorrhage detection competition dataset
kaggle competitions download -c rsna-intracranial-hemorrhage-detection
mkdir data && unzip -o rsna-intracranial-hemorrhage-detection.zip -d data
  1. Show the data contents
ls data/rsna-intracranial-hemorrhage-detection/
# stage_2_sample_submission.csv  stage_2_test  stage_2_train  stage_2_train.csv
du -h data/rsna-intracranial-hemorrhage-detection/
# 60G     data/rsna-intracranial-hemorrhage-detection/stage_2_test
# 371G    data/rsna-intracranial-hemorrhage-detection/stage_2_train
# 431G    data/rsna-intracranial-hemorrhage-detection/
  1. Run the process_and_view_dicom Jupyter notebook.