Detect Parkinson's disease just from a wave sketch image!
Parkinson's disease is a brain disorder that causes unintended or uncontrollable movements, such as shaking, stiffness, and difficulty with balance and coordination. Traditionally, to detect Parkinson MRI images are used.
In 2017 a paper was published to detect Parkinson's by using hand sketch images. They proposed a dataset that includes wave and spiral sketches from healthy and Parkinson's people. And they said we can use these sketch images to detect Parkinson's disease!
I provided a method to detect Parkinson's based on the proposed dataset in paper.
You can access the dataset from Kaggle website.
Run this notebook on Google Colab and test on the proposed dataset.
You can download the pretrained model here
Accuracy on train set: 100%
Accuracy on test set: 96.~%
