DiagnoX – A New Way to Diagnose Rare Diseases

DiagnoX is an open-source project dedicated to diagnosing diseases. The original code is designed and geared towards Aortitis, a rare cardiovascular disease characterized by inflammation in the aorta. Whether you are searching for inspiration for a personal project, a new hobby, or you just want to give back to the community, DiagnoX is a great way to start.

To store and compare the CT scans, the python program uses HOG (histogram of oriented gradients) descriptors. They store a histogram of gradients (consisting of x and y derivatives which have direction and magnitude), which is more efficient than storing the entire image because the “useful” data consists of abrupt changes in the derivatives.

For classification, Linear-SVC, an algorithm which establishes a hyperplane between clusters of data, is used. LinearSVC uses the parameters which the HOG descriptor provides to train the program and draw the hyperplane, effectively classifying each image as either having or not having an inflamed aorta.

With an overall accuracy rate of 94% and a type II (false negative) error rate of only 1.4%, the algorithm proves to be effective.

If you would like to join this community of developers by creating efficient algorithms to diagnose rare diseases, please click here.



3 thoughts on “DiagnoX – A New Way to Diagnose Rare Diseases”

    1. Curious if you tried deep learning and what accuracy you got from it vs the methods you mention above. Will clone your github repo and try it with my data sometime. Thanks!

      Liked by 1 person

      1. Yes! I did try a neural network (a Yann-Lecun Model, which only has one hidden layer). However, due to the lack of data of Aortitis CT scans, its accuracy was quite low, even after augmentation.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s