adhi1904/pds

Parkinson detection based on wave sketches and deep learning - PDS

Parkinson detection based on sketches and deep learning - PDS

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 published to detect Parkinson by using hand sketch images. They proposed a dataset that includes wave and spiral sketches from healthy and parkinson person. And they said we can use these sketch images to detect Parkinson's disease!

About this project

I provied a method by using Few-Shot learning to detect Parkinson based on the proposed dataset in paper.

Dataset

You can access to dataset from Kaggle website.

Notebook

Run this notebook on Google Colab and test on proposed dataset.

Results

You can download the pretrained model here

Accuracy on train set: 100%

Accuracy on test set: 96.~%

Dataset samples

License

MIT License

Based on

• Easy Few-Shot Learning