Wearable sensor trained to count cough

There are many problems that are easy for humans to solve, but nearly impossible for computers to solve. Even though it seems like modern computing power being what it is, we should be able to solve many of these problems, things like identifying objects in images are still quite difficult. Likewise, identifying specific sounds in audio samples remains problematic and, as [Eivind] discovered, sets back much medical research. To solve a specific problem, he created a medical patient cough counting system.

This was built with the idea of ​​helping people with Chronic Obstructive Pulmonary Disease (COPD). Most existing methods for studying the disease and treating patients with it involve manually counting the number of coughs on an audio recording. Although there are software solutions to this problem to save time, this device seeks to identify coughs in real time as they occur. It does this by training a model using tinyML to identify cough and reject cough-like sounds. Everything runs on an Arduino Nano with BLE for communication.

Although the only data the model was trained on is the sounds of [Eivind], the existing prototypes look promising. With more solid data, this could be a powerful tool for patients with this condition. And, even though this uses machine learning on a small platform, we've already seen that Arudinos is quite capable of being effective machine learning solutions with the right tools on board.