ECG as Tensor
Re-framing the ECG signal as a tensor so we can be done with “palm readings”.
The electrocardiogram is New DBS hardware lets us record from the implanted DBS lead. Being able to quickly assess the quality of the recordings at the patient’s bedside is lets us maximize the signal-to-noise in the recordings. I developed a MATLAB based UI for LFP spectrogram assessment directly at the patient’s bedside.
Why is this important?
To be clear, this isn’t anything new. There has been a lot of excellent work done on analysing ECGs with modern tools, including cutting edge approaches in machine learning (ML). What seems to have always been lacking is an understanding on the clinical side of why these math-based approaches are better and must be adopted ASAP.
Clinicians using DBS are starting to listen on on brain activity as they try to optimize stimulation for various brain disorders. Having an easy-to-use tool that enables high-level analysis of recordings can go a long way to ensuring the success of a research proposal.