Skip to content

Latest commit

 

History

History
22 lines (16 loc) · 874 Bytes

README.md

File metadata and controls

22 lines (16 loc) · 874 Bytes
Raw

Research on carina and left hilum detection for COVID19 patients in CXR

This source code is for sharing deep learning-based detection model for carina and left hilium in CXR. It has been validated for COVID19 patients as well.

Lung segmentation in CXR is followed by this detection model for dividing the whole lung into four regions.

Pre-requisities

  • Python == 3.5
  • Keras >= 2.0
  • Tensorflow >= 1.11
  • Scipy
  • OpenCV
  • keras-retinanet

Detection result and segmentation for four regions (mAP: 0.69 for a private dataset)

Detection

Reference (Github)

  1. https://github.com/qubvel/segmentation_models
  2. https://github.com/fizyr/keras-retinanet
  3. https://github.com/younggon2/Research-Segmentation-Lung-CXR-COVID19