MIT Solve

Solution Overview

Solution Name

Nung

One-line solution summary.

Improved medication accuracy for Parkinson's Disease patients

Elevator pitch

What is your solution?

Nung is a wearable tech device powered by machine learning regression model to track and analyze Parkinson's Disease patients’ vibration patterns, identify the exact time when they feels the most vulnerable and recommend medication taking-time to doctors.

What specific problem are you trying to solve?

More than 10 million people worldwide are living with Parkinson’s Disease, a progressive nervous system disorder that occurs when brain cells which produce dopamine stopped working or died.

Generally, Parkinson’s Disease is categorized into 2 symptoms. Motor symptoms: resting tremor, bradykinesia, slowness of movement etc, and non-motor symptoms: memory or cognitive problems, speaking in a soft monotone voice, sleep difficulties etc.

One of the hardest things about Parkinson's Disease is there is no "one-size" fits all approach, the symptoms and progression from patient to patient varys a lot. Dr Rachel Dolhun, a movement disorder specialist said “if you met one person with Parkinson’s, you’ve met one person with Parkinson’s.”

Since, Parkinson's Disease is one of the non-curable disease, doctors could only recommend medication and treatments to alleviate patients' symptoms. As a result, the demand for better and more accurate medication is increasing.

Nung solves this issue by understanding the underlying tremor pattern in patients day-to-day lives, finding the tremor peaks throughout ones day, and recommending doctors patients' medication-taking time to help more effectively cope with their effects of the illness.

Who does your solution serve? In what ways will the solution impact their lives?

My grandmother had Parkinson’s Disease for more than a decade. Having experienced her symptoms' progression over time first-handly, my family and I faced countless challenges when taking care of her.

Every six months, my grandma would visit the hospital for a check-up and report her conditions to the doctor. However, she couldn’t accurately pinpoint her condition as she has difficulties remembering her symptoms overtime. This vague description led to imprecise medication that targeted the wrong symptoms.

Nung serves as a bridge between patient and doctor, the final result -- the time the patient feels the most vulnerable in 30 minutes interval -- provides an accurate summary of the patient’s past and current symptoms to doctors. Helping doctors to better offer more personalized medication to each Parkinson's Disease patient. With little Parkinson’s Disease vibration data available online, Nung could act as a database that holds records from different stages of Parkinson’s Disease patients. Furthermore, allowing researchers to further understand and analyze in future medication development.

What steps have you taken to understand the needs of the population you want to serve?

Tested Nung on 3 Parkinson's Disease patients.

Methods:

A Parkinson’s Disease patient wears Nung on their wrist
Connect Nung with their home WiFi or mobile phone hotspot
The SW420, vibration module, records the vibration value of the patient during the day for 8 consecutive hours
Calculate the average vibration value every 5 minutes
Group six 5 minutes vibration value together
Calculate the average vibration value every 30 minutes
Output the calculated data to Firebase: cloud database
Fetch the data from Firebase and display it on a React.js website to show patients their daily symptom pattern
Download the data from Firebase in CSV format
Feed the CSV data to the machine learning algorithm
Sketch a graph with all the data points and the curve of best fit
Identify the top 5 time the patient vibrates the most vigorously throughout the day

Mentored by Hong Kong City University's Dr. Ray Cheung and Hong Kong Polytechnic University's Dr. Margaret Mak, a council member of the Hong Kong Parkinson’s Disease Foundation.

Which aspects of the Challenge does your solution most closely address?

Improving healthcare access and health outcomes; and reducing and ultimately eliminating health disparities (Health)

Our solution's stage of development:

Prototype: A venture or organization building and testing its product, service, or business model

Explain why you selected this stage of development for your solution—in other words, what have you accomplished to date?

I only reached out to 3 Parkinson's Disease patients around my community to test out my prototype device.

I believe in order to achieve personalized medication for every Parkinson’s Disease patient, a better machine learning model must be implemented that not only accounts for several hundreds of vibration pattern, but also medication options patients take to alleviate Parkinson’s Disease symptoms.

Where our solution team is headquartered or located:

Hong Kong

Team Lead:

Melanie Hsieh