Knee Osteoarthritis Detection and Severity Prediction

Download Project Document/Synopsis

Nevon Knee Osteoarthritis Detection Python
Tested
nevon software

Arthritis is a disorder that causes swelling, tenderness, inflammation, stiffness etc. in one or more joints. Arthritis is more common in older people and typically worsens with age. While there are many different types of arthritis with different causes and treatments, osteoarthritis is the most prevalent. Osteoarthritis is estimated to affect nearly 237 million people globally, this accounts for almost 3.3% of the human population. Although as of now there is no known cure for arthritis, the benefits of early detection can’t be understated.

The Knee Osteoarthritis Detection helps patients detect Osteoarthritis in their knees. Along with helping with early detection, this web application also detects the severity of the disorder. The patient has the option to register and fill his details. He has to give his name, username, x-ray photo, guardian photo, his own photo, identity photo, licence photo and other details. We can have people of same names but we cannot have people with similar username. And hence there will be no mis-match. Each person will have a unique username and a particular detection along with its symptoms and solution.

We first train the datasets. We have 5 classes of images and after training by CNN algorithm used in this project, we have different convolutional layers forming on it and after each layer, the precision keeps increases. How many layers being used depends upon the number of eposch used. More the number of eposch, more the precision. After a .h5 file is created which contains all the attributes and confidence score for each detection class.

Once the x-ray photo has been uploaded, the photo undergoes resizing and color change to grey and then there several convolutional layers applied on it by the help of CNN algorithm used in this project and a confidence score is generated and is matched with the confidence score in the .h5 file and the class is detected and hence detection is done

Advantages

  • Helps with instantly detecting signs of Osteoarthritis in knees.
  • This system even informs about the severity of arthritis.
  • Easy to use.

Limitations

  • Wrong inputs will affect the project outputs.
-->