Robust Unsupervised Transient Detection With Invariant Representation based on the Scattering Network

TL;DR

This paper introduces a robust, unsupervised method for transient detection in noisy signals using a scattering network to achieve frequency-invariant representations, with applications in epileptic seizure onset prediction.

Contribution

It develops a novel frequency-invariant scattering network approach for unsupervised transient detection, improving robustness against frequency shifts and noise.

Findings

Effective seizure onset prediction from subdural recordings.

Robust detection of inter-ictal spikes in noisy environments.

Low asymptotic complexity of the proposed method.

Abstract

We present a sparse and invariant representation with low asymptotic complexity for robust unsupervised transient and onset zone detection in noisy environments. This unsupervised approach is based on wavelet transforms and leverages the scattering network from Mallat et al. by deriving frequency invariance. This frequency invariance is a key concept to enforce robust representations of transients in presence of possible frequency shifts and perturbations occurring in the original signal. Implementation details as well as complexity analysis are provided in addition of the theoretical framework and the invariance properties. In this work, our primary application consists of predicting the onset of seizure in epileptic patients from subdural recordings as well as detecting inter-ictal spikes.