Attention Autoencoder for Generative Latent Representational Learning in Anomaly Detection

Abstract

Today, accurate and automated abnormality diagnosis and identification have become of paramount importance as they are involved in many critical and life-saving scenarios. To accomplish such frontiers, we propose three artificial intelligence models through the application of deep learning algorithms to analyze and detect anomalies in human heartbeat signals. The three proposed models include an attention autoencoder that maps input data to a lower-dimensional latent representation with maximum feature retention, and a reconstruction decoder with minimum remodeling loss. The autoencoder has an embedded attention module at the bottleneck to learn the salient activations of the encoded distribution. Additionally, a variational autoencoder (VAE) and a long short-term memory (LSTM) network is designed to learn the Gaussian distribution of the generative reconstruction and time-series sequential data analysis. The three proposed models displayed outstanding ability to detect anomalies on the evaluated five thousand electrocardiogram (ECG5000) signals with 99% accuracy and 99.3% precision score in detecting healthy heartbeats from patients with severe congestive heart failure.

Reservoir based spiking models for univariate Time Series Classification

“…We therefore group all the 4 abnormal classes into one class. Thus, the ECG5000 time-series classification task is modeled as a binary classification task between healthy and unhealthy heartbeats; a few authors do the same (Matias et al, 2021 ; Oluwasanmi et al, 2021 ; Biloborodova et al, 2022 ).…”

Exploration of Sleep Events in the Latent Space of Variational Autoencoders on a Breath-by-Breath Basis

“…Outside medical approaches, VAE has been used for tasks such as anomaly detection (An & Cho, 2015), text classification (Xu, Sun, Deng, & Tan, 2017), and recommender systems (Li & She, 2017). In the medical field, VAE and other AE methods have been applied in various ways, mainly for the electrocardiogram (ECG) (Costa, Sánchez, & Couso, 2021;Kuznetsov, Moskalenko, Gribanov, & Zolotykh, 2021;Oluwasanmi et al, 2021), but also with some applications for the electromyogram (EMG) and EEG (Singh & Ogunfunmi, 2021). The application of VAE and other generative machine learning models for biomedical signal analysis is a largely unexplored field.…”

Research on hybrid transformer-based autoencoders for user biometric verification

“…In contrast, in [14], adversarial autoencoder [15], which is the combination of autoencoder with generative adversarial networks (GAN), was used for the health monitoring of ECG and for detecting abnormal data points, which by the authors outperformed LSTM and VAE architectures. The autoencoder with attention mechanism, placed between encoder and decoder blocks to learn relations on the latent space feature representations, is proposed in [16] for ECG data anomaly detection.…”