Search

Real-world anomaly detection for time series is still a challenging task. This is especially true for periodic or quasi-periodic time series since automated approaches have to learn long-term correlations before they are able to detect anomalies. Electrocardiography (ECG) time series, a prominent real-world example of quasi-periodic signals, are investigated in this work. Anomaly detection algorithms often have the additional goal to identify anomalies in an unsupervised manner. In this paper we present an unsupervised time series anomaly detection algorithm. It learns with recurrent Long Short-Term Memory (LSTM) networks to predict the normal time series behavior. The prediction error on several prediction horizons is used to build a statistical model of normal behavior. We propose new methods that are essential for a successful model-building process and for a high signal-to-noise-ratio. We apply our method to the well-known MIT-BIH ECG data set and present first results. We obtain a good recall of anomalies while having a very low false alarm rate (FPR) in a fully unsupervised procedure. We compare also with other anomaly detectors (NuPic, ADVec) from the state-of-the-art.

When data are transmitted between electronic patient record (EPR) systems, we can distinguish several tasks. One task is the definition of structure and semantic content of the data in a message structure. Another task is the mapping of the sending EPR's structure to this message structure. A third task is the mapping of the message structure to the receiving EPR's structure. We describe an approach, which distinguishes clearly between these different tasks and activities. Using this approach we have implemented a data transfer procedure between a cancer registry application and a middleware for healthcare information systems. Our experience showed that the proposed systematic approach helped identify problems for data transfer in an early design phase. It also allowed us to limit modifications of the data exchange procedure to certain tasks or activities when one of the EPR applications was updated. In the end, we could even exchange the underlying message format without having to reimplement the complete interface.