Mini-project
Electronics, Semiconductors & Telecom
ECG heartbeat classifier for a wearable
Classify heartbeats from ECG signals, small enough to run on a wearable.
The brief
A start-up building a low-cost heart monitor needs a model that sorts each heartbeat into normal or one of several arrhythmia types, and is small enough to run on a microcontroller. Good looks like: good accuracy on the rare beat types, a model under a few hundred kilobytes, and a note on how it would run on the device.
What you will build
- Explore and visualise the beat classes
- Handle the class imbalance
- Train a 1-D CNN and a small classical baseline
- Evaluate per class with a confusion matrix
- Shrink the model (pruning or quantisation) and report its size and speed
Free, open data to build it with
Check each dataset's licence before you publish your work. Your mentor gives your team its own version of the brief (data, scope or region), so every team's project is its own.