Related Experiment Videos
Wavelet-based event detection in implantable cardiac rhythm management devices
Magnus Aström1, Salvador Olmos, Leif Sörnmo
1Department of Electrical Engineering, Lund University, SE-221 00 Lund, Sweden.
IEEE Transactions on Bio-Medical Engineering
|March 15, 2006
Summary
A new event detector for implantable devices uses wavelet transforms and a likelihood ratio test for reliable signal detection. This algorithm achieves high accuracy even with common noise sources at intermediate signal-to-noise ratios.
Area of Science:
- Biomedical Engineering
- Signal Processing
- Implantable Devices
Background:
- Accurate event detection is crucial for implantable medical devices.
- Existing methods may struggle with noise and require high computational power.
Purpose of the Study:
- To develop a novel, low-complexity event detector for implantable devices.
- To evaluate the detector's performance against various noise sources and signal-to-noise ratios.
Main Methods:
- A signal model representing events as linear combinations of basis functions.
- Utilized the dyadic wavelet transform with integer filter coefficients.
- Employed a generalized likelihood ratio test for detection.
Main Results:
- Reliable event detection achieved at a signal-to-noise ratio (SNR) of 25 dB.
- The detector demonstrated robust performance against diverse common noise types.
- Achieved a low overall performance of 0.7% missed events and 0.1% false alarms.
Conclusions:
- The proposed wavelet-based detector offers an efficient and accurate solution for implantable devices.
- The algorithm provides reliable electrogram analysis in the presence of noise.
- This technology has the potential to improve the functionality of implantable monitoring systems.