Pneumonia is a form of acute respiratory infection that affects the lungs. The lungs are made up of small sacs called alveoli, which fill with air when a healthy person breathes. When an individual has pneumonia, the alveoli are filled with pus and fluid, which makes breathing painful and limits oxygen intake.
An X-ray helps your doctor look for signs of inflammation in your chest. If inflammation is present, the X-ray can also inform your doctor about its location and extent.
Pneumonia-Detector attempts to automate methods to detect and classify pneumonia from medical x-ray images using a Convolutional Neural Network. It is able to detect correctly 88% of pneumonia cases but it is NOT in any way a substitute for consulting a professional medical examiner.
Click to view full architecture
│ training.ipynb (v2)
│ inference.ipynb (v2)
│ detect-voila.ipynb (v1)
│ script.py (v1)
│ pneumonia-detection.ipynb (v1)
│
├───utils (v1)
│ ├──constants.py (v1)
│ ├──pneumonia_model.py (v1)
│ └───utils.py (v1)
│
└───weights
└─── pneumonia_detector_model.pth (v1)
For coders: Use the 'diagnose' method in script.py either by importing or editing the script file itself. Pass an x-ray image (either a PIL.Image, torch.tensor, numpy.array or even a path to the image file) as argument to the function.
For non-coders: Visit this Binder link, wait for it to render, sip some coffee as you wait :).
Visit the Colab notebook by clicking here
Visit Binder to try it yourself.
Placeholder | Prediction |
---|---|
Visit the Colab notebook by clicking here and interact with the Gradio Interface
Dahir Ibrahim (Deedax Inc) - http://instagram.com/deedax_inc
Email - [email protected]
YouTube - https://www.youtube.com/@deedaxinc
Twitter - https://twitter.com/DeedaxInc
Project Link - https://github.com/Daheer/Pneumonia-Detection