Frankenstein

Frankenstein: a simple monitoring solution in a world of distractions

Comment 1

The idea for this project emerged from a disappointing start to SpeedHacks 2.0, where my original teammates decided to skip the competition. Faced with a tight deadline, I struggled to find a team that would accept a beginner with Arduino hardware. Just as I was about to give up, I found a teammate who pointed out changes in my breathing when I was taking the stairs. He linked these symptoms to anxiety, and we both realized that this observation could lead to an interesting project idea.

Together, we decided to create an offline device that could monitor anxiety symptoms for patients who have limited access to medical professionals. The device would track key physiological and environmental factors related to anxiety, like sound (which could detect changes in breathing), and atmospheric values (temperature, humidity, and air pressure). At a later stage, we planned to add an acceleration sensor to detect shakiness in patients, as that too can be linked to anxiety symptoms.

To make this project even more impactful, we incorporated AI into the design. The AI was trained in Python using data collected from the Arduino hardware. The data was compared and analyzed to determine if certain thresholds of environmental data or sound patterns matched those typically seen in anxious patients. A system was also built to log the patient’s symptoms with timestamps, allowing patients to track their condition over time. This could serve as a useful tool for those who are unable to see a doctor regularly but still want to keep track of their health and provide their recorded symptoms when they are ready to seek medical attention.

In essence, this project aims to empower individuals by providing a tool that helps them monitor their mental health in an accessible, affordable, and efficient manner.

Technologies Used in the Project ->Arduino IDE: For writing and uploading the code to the Arduino UNO R3. ->Python: For data collection, machine learning model training, and data visualization. ->Machine Learning (Scikit-learn): For training the decision tree model to classify anxiety based on sensor data. ->CSV: For storing sensor data and model results. ->Jupyter Notebooks: For visualizing and analyzing the sensor data. ->GitHub: For version control and collaboration on the project.

Arduino Libraries: ->Arduino_SensorKit: For interacting with the sensor kit on the Arduino (including temperature, humidity, and pressure sensors). ->DHT Library: For working with the DHT11 temperature and humidity sensor. ->Wire: For I2C communication with the OLED display. ->Decision Trees: A machine learning algorithm used to detect patterns in the sensor data that are indicative of anxiety.

Built With

Share this project:

Updates