Calculated respiratory rate
Peak detection algorithm
Respiratory diseases are widespread across the global population. The most serious examples include chronic obstructive pulmonary disease (COPD), asthma and sleep apnea; all affecting a large amount of patients and leading to thousands of patient deaths annually. In many cases, the continuous monitoring and quick diagnosis of respiratory vital signs could be the determining factor to save a life.
With a lot of the technological advances being dedicated into making portable ECG monitors, it has been detected by physicians and engineers alike that respiratory rate and other lung-related biosignals have been neglected within the clinical setting.
Respiratory rate has been reported to be a useful metric, not only for critically ill patients but also for overall health tracking - athletes, patients undergoing recovery, etc. The close monitoring of neo-natal premature infants is yet another medical scenario that would benefit from modern-day technology, as it presents a practical manner to detect breathing anomalies, potentially preventing deathly diseases (eg. sudden infant death syndrome).
Our project aims to fill the technological void in the monitoring of respiratory signals.
What it does:
The device measures the acceleration of chest movements and sends this wirelessly to a smartphone app. After being processed, the data can then be visualized as a time series, or as an averaged respiratory rate.
How we built it:
Used existing, open-source libraries for the hardware components (accelerometer and bluetooth module). For the data processing, implemented a smooth z-wave algorithm to detect the peaks of the breathing signal. For the app development, used Swift within XCode.
Challenges we ran into:
Creating a functional workflow among the hardware components: sensors, microprocessor and Bluetooth module. Learning to implement data processing functions within Swift (as opposed to more popular choices such as Python or MATLAB). Sharing data across different storyboards in an iOS app.
Accomplishments that we're proud of:
A continuous measurement of respiratory rate with an easy-to-use, portable device. The fact it connects to a smartphone app opens up a wide range of future applications to be explored (health tracking, assisting GP practices, enhancing clinical monitoring, etc).
What we learned:
App development, data analysis in Python and Swift, embedded programming and its interface with Swift.
What's next for Respi-mate:
Rely solely on the smartphone’s sensors for measuring chest movements. Further exploration of the output from accelerometers to provide more insights about patient data. Enhance UI and UX.