Semester Embedded Systems Challenge
Over 1 million people in the USA and more than 10 million people worldwide suffer from Parkinson’s disease. A critical challenge in managing Parkinson’s is the clinically accurate detection of symptoms to enable therapy optimization.
One of the most common symptoms is the resting tremor, which affects over 70% of patients. Resting tremors occur when a body part (typically the hand or wrist) is completely supported and at rest, and they are minimal or absent during voluntary activity.
The classical Parkinsonian tremor typically occurs at 3 to 6 Hz (cycles per second).
This project focuses on developing a wearable Parkinsonian tremor detector using only the STM32F429 Discovery board with its embedded gyroscope (L3GD20).
The goal of this challenge is to:
✅ Capture real-time rotation data using the onboard gyroscope by measuring angular velocities.
✅ Analyze time-segmented data to detect tremor patterns within the target frequency range (3-6 Hz).
✅ Provide a visual indication of the presence and intensity of resting tremors using available board resources (e.g., LEDs, LCD screen).
- No additional hardware is allowed—use only the STM32F429 Discovery board and its built-in sensors and display resources.
- Tremor Detection: Identify the presence of resting tremors based on frequency analysis.
- Intensity Estimation: Indicate the severity or strength of the tremor.
- Real-time Feedback: Display visual cues for tremor status using the board's LEDs or LCD.
- STM32F429 Discovery Board
- Onboard Gyroscope (L3GD20)
- C programming & embedded systems knowledge
- Signal processing techniques (e.g., FFT, filtering)
- This challenge emphasizes embedded programming, signal analysis, and real-time processing.
- Explore creative ways to visualize tremor data using the limited resources of the board.
- Consider using filtering techniques to isolate the target frequency range (3-6 Hz) from noise.
Let’s build a meaningful solution that could make a real difference in Parkinson’s disease management!