Park-Detect

Inspiration

Last year, Someone close to me back home broke down the news to me that she had been diagnosed with Parkinson's. We have spent more than 20-30k just on its detection, medical fees (Excluding doctor's appointment). I want to create this because I do not want others to go through what we are going. It is a simple Project, yet effective.

What it does

Detect Early signs of Parkinson's Disease

How I built it

I created a cluster of RP. One Master and One slave. Create a Cluster > Data > Clean it Up > Used mpi4py on Cluster Trained SVC (Support Vector Classification) classifier using 11 features by scikit-learn. In the server, the Master Node was used as an API of cherrypy which determines whether the patient has Parkinson’s or not. Later on, the Slave node is used to process new model with a complete data will create new model that will sync with master depending upon certain threshold set and usage of master.

Challenges I ran into

Everything can crash and stop working at any point of time. You just have to be patient

Accomplishments that I'm proud of

I'm proud of not attempting to incorporate any of the sponsor companies' APIs just to be eligible for their prizes because that would have caused unnecessary complexity and distracted me from achieving the main goal.

What I learned

To Prove People Wrong

What's next for Park-Detect

I do not create projects at hackathons to win, I create them to make actual products. Each and every product I have created at a hackathon have used somewhere or the other

Built With

• cherrypy
• html
• javascript
• mpi4py
• pandas
• python
• raspberry-pi
• scikit-learn