Semi-supervised Stacked Label Consistent Autoencoder for Reconstruction and Analysis of Biomedical Signals

TL;DR

This paper introduces a semi-supervised autoencoder framework that simultaneously reconstructs and classifies biomedical signals, outperforming traditional methods in speed and accuracy for ECG and EEG analysis.

Contribution

It presents the first combined autoencoder-based approach for reconstruction and classification of biomedical signals, replacing traditional compressed sensing techniques.

Findings

Reconstruction is over an order of magnitude faster than CS-based methods.

The method achieves superior classification accuracy.

Operates effectively in a semi-supervised setting.

Abstract

In this work we propose an autoencoder based framework for simultaneous reconstruction and classification of biomedical signals. Previously these two tasks, reconstruction and classification were treated as separate problems. This is the first work to propose a combined framework to address the issue in a holistic fashion. Reconstruction techniques for biomedical signals for tele-monitoring are largely based on compressed sensing (CS) based method, these are designed techniques where the reconstruction formulation is based on some assumption regarding the signal. In this work, we propose a new paradigm for reconstruction we learn to reconstruct. An autoencoder can be trained for the same. But since the final goal is to analyze classify the signal we learn a linear classification map inside the autoencoder. The ensuing optimization problem is solved using the Split Bregman technique. Experiments have been carried out on reconstruction and classification of ECG arrhythmia classification and EEG seizure classification signals. Our proposed tool is capable of operating in a semi-supervised fashion. We show that our proposed method is better and more than an order magnitude faster in reconstruction than CS based methods; it is capable of real-time operation. Our method is also better than recently proposed classification methods. Significance: This is the first work offering an alternative to CS based reconstruction. It also shows that representation learning can yield better results than hand-crafted features for signal analysis.