Real-time parkinsonian tremor detection system

• 

  Code environment - Platform IO VSCode

Inspiration

Since my best friend is studying to become a Doctor,I'm particularly interested in healthcase based projects. The idea for this project arose from a desire to create a practical tool that could assist in the early detection and monitoring of Parkinson's disease! As we know, Parkinsonian tremors are a key symptom of this condition, and being able to detect and analyze these tremors in real-time can greatly aid in diagnosis and management. .

What it does

This system utilizes a gyroscope to detect tremors in individuals suspected of having Parkinson's disease. It records angular velocities, calculates the root mean square (RMS) of these movements, and determines whether the movements indicate a tremor based on predefined thresholds. The system provides real-time feedback on the detected tremors and their intensity, logs the data, and displays the information on the board's LCD screen.

How we built it

I developed the system using an STM32F429IDISCOVERY development board and its peripherals. The project was built on PlatformIO using the Mbed framework, with code written in embedded C. Key components include an SPI gyroscope for motion detection, an LCD for data display, a push-button for user interaction, and onboard LEDs for visual feedback.

Challenges we ran into

One of the main challenges was configuring the gyroscope to read data correctly. This required extensive study of the SPI-based architecture and the datasheet of the STM32F429IDISCOVERY board. Ensuring accurate and reliable data collection from the gyroscope was critical to the project's success!!

Accomplishments that we're proud of

I am proud of my innovative approach to tremor detection, which involved using real-time graphs and individual-specific thresholds rather than a generalized FFT analysis for frequency detection. This method allows for more personalized and accurate detection of tremors, improving the system's overall effectiveness.

What we learned

Throughout this project, I gained a deeper understanding of embedded systems, SPI communication, and real-time data processing and also about what Parkinsonian tremors are like! I also learned about the challenges of designing medical monitoring systems and the importance of tailoring solutions to individual needs, which took my project to a whole new level.

What's next for Real-time Parkinsonian Tremor Detection System

My next steps include creating a more portable device. I also plan to integrate IoT capabilities to enable remote monitoring and data analysis. This would allow healthcare providers, doctors or family to track patients' tremor data in real-time, providing more comprehensive and timely care.

This would be a great idea to implement on a car's steering wheel too. If a patient has Parkinson's disease, there's a high possibility of the tremors getting too bad whilst he/she is driving, in which case the detector can send a signal to the car brakes and also send out alerts to their emergency contacts. (Electric Vehicles only).

Keep in mind, this particular logic and analysis are written particularly for Parkinsonian tremors because they follow a specific pattern. Only analysing frequency of movements in all directions is the easier way out and has been done before, but this project aims to have higher accuracy and is focused on Parkinson's disease particularly.

Built With

• c
• c++
• mbed
• platformio
• stm32