PDGAN: A Early Detection System for Parkinson's Disease Powered by Deep Learning
PDGAN allows for the accurate, quick, and minimally invasive diagnosis of Parkinson's Disease.
Are you employed by T-Mobile or related to an employee of T-Mobile?
Eligibility: Date of Birth
February 2nd, 2002
Help us stay in touch!
Phone Number: 5712636832
Address: 3123 Oxford Forest Drive Virginia: Herndon (20171)
Website or social media url(s) (optional):
Date You Started Your Project Started
Project Stage: Select the description below that best applies to your approach.
Established (successfully passed early phases, have a plan for the future)
1. The Problem: What problem are you helping to solve?
Currently, Parkinson's Disease afflicts millions of lives, but is very hard to diagnose in its earliest stages. This is because the disease can only be detected through symptoms including stiffness of movement, rigidity, and others. Along with this, the diagnosis accuracy for Parkinson's is currently around 80%, which means that many patients are not diagnosed, and those that are are diagnosed too late for most treatment to be effective.
2. Your Solution: How are you planning to solve this problem? Share your specific approach.
PDGAN, a Deep Learning-powered end-to-end system for the early detection of Parkinson's Disease, is looking to find early signs of the disease through the MRI scan, which looks at structural changes in the brain. I believe that because Parkinson's Disease is a neurological disorder, the root cause should be found in the brain. There, I used imagine analysis techniques like Convolution Neural Networks to create a predictive model capable of diagnosing the disease in a fraction of the time.
The system utilizes many computational optimizations and tricks to perform better and quicker compared to traditional models, including a generative component that is able to synthesize fake (but lifelike) MRI scans of "fake" patients.
3. Personal Journey: What’s the story behind why you decided to start this project?
Around two years ago, my grandfather was diagnosed with Parkinson's Disease. His diagnosis and the future news that most treatment options would not be effective because he was diagnosed at a late stage was what drove me to begin to do further research into Parkinson's Disease. After doing so, I realized that millions of people face the same problem, and that Parkinson's Disease is more widespread than I originally thought.
4. Selfie Elevator Pitch: Include 1-minute video that answers the following “I am stepping up to make change because...”
5. Example: Please walk us through a specific example of what happens when a person or group gets involved with your project.
As a patient comes to terms with potential early signs of Parkinson's Disease, they would go to the neurologist's office to get an initial checkup (this happens often). Instead of being deferred to come back at a later time for a more accurate diagnosis, the neurologist that partners with PDGAN would be able to take an MRI scan and get a cheap and quick diagnosis right then and there.
6. The X Factor: What is different about your project compared to other programs or solutions already out there?
This project is different compared to others because of the power that it has in generalization. Not only does it solve most of the problems faced by other research projects through computational tricks (such as the generation of synthetic MRI images), but this system is also generalizable to other neurological disorders, such as Alzheimer's Disease. Through the same methodology and a different dataset, the same impact that PDGAN has with Parkinson's Disease can be achieved with other diseases.
7. Impact: How has your project made a difference so far?
Although PDGAN hasn't been deployed (as there are many medical background checks that need to be in place before that is the case), the design of PDGAN has been provisionally patented and I am in talks with the Michael J Fox Foundation (a leading nonprofit organization in Parkinson's Research) to get the research and system in clinical trials for future potential impact. Then, I would partner with local doctors to get a pilot program started, then hopefully push nationally and internationally with this design. Because this is completely electronic, there is no physical apparatus necessary, and the sending of information would be instantaneous.
8. What’s Next: What are your ideas for taking your project to the next level?
Right now, I am expanding this research to other biomarkers of Parkinson's Disease to make a more comprehensive model for Parkinson's Disease and the diagnosis of it at an early stage. I am also in talks with the Michael J Fox Foundation (a leading nonprofit organization in Parkinson's Research) to get the research and system in clinical trials for future potential impact. Currently, I have been awarded as a Davidson Research Fellow and along with a research scholarship of $25,000 I have access to a network of researchers and peers who I can talk to and get more information about the research to development process.
9. Which of the following types of expertise would be most useful for you?
10. Finances: If applicable, have you mobilized any of the following resources so far?
Help Us Support Diversity! Part 1 [optional] Which of the following categories do you identify with?
Asian (for example: Chinese, Filipino, Indian, Vietnamese, Korean, Japanese, Pakistani) (9)
Help Us Support Diversity! Part 2 [optional] Do you identify as part of any of the following underrepresented communities?
No, I do not identify with an underrepresented community
How did you hear about this challenge?
Participated in previous Ashoka challenges
Ashoka page or contact
Referral: If you discovered the Challenge thanks to an organization or person other than Ashoka or T-Mobile, who was it?