In many real-world applications today, it is critical to continuously record and monitor certain machine or system health indicators to discover malfunctions or other abnormal behavior at an early... Show moreIn many real-world applications today, it is critical to continuously record and monitor certain machine or system health indicators to discover malfunctions or other abnormal behavior at an early stage and prevent potential harm. The demand for such reliable monitoring systems is expected to increase in the coming years. Particularly in the industrial context, in the course of ongoing digitization, it is becoming increasingly important to analyze growing volumes of data in an automated manner using state-of-the-art algorithms. In many practical applications, one has to deal with temporal data in the form of data streams or time series. The problem of detecting unusual (or anomalous) behavior in time series is commonly referred to as time series anomaly detection. Anomalies are events observed in the data that do not conform to the normal or expected behavior when viewed in their temporal context.This thesis focuses on unsupervised machine learning algorithms for anomaly detection in time series. In an unsupervised learning setup, a model attempts to learn the normal behavior in a time series — which might already be contaminated with anomalies — without any external assistance. The model can then use its learned notion of normality to detect anomalous events. Show less
'Cancer stem cells' (CSCs) are tumor cells with stem cell properties hypothesized to be responsible for tumorigenesis, metastatis, and resistance to treatment, and have been identified in different... Show more'Cancer stem cells' (CSCs) are tumor cells with stem cell properties hypothesized to be responsible for tumorigenesis, metastatis, and resistance to treatment, and have been identified in different tumors including cutaneous melanoma, using stem cell markers such as CD133. This study explored expression of CD133 and other putative stem cell markers in uveal melanoma. Eight uveal melanoma cell lines were subjected to flow-cytometric (fluorescence-activated cell sorting) analysis of CD133 and other stem cell markers. Eight paraffin-embedded tumors were analyzed by immunohistochemistry for CD133, Pax6, Musashi, nestin, Sox2, ABCB5, and CD68 expressions. Ocular, uveal melanoma, and hematopoietic stem cell distributions of C-terminal and N-terminal CD133 mRNA splice variants were compared by reverse-transcription PCR. Fluorescence-activated cell sorting analysis revealed a population of CD133-positive/nestin-positive cells in cell lines Mel270, OMM 2.3, and OMM2.5. All cell lines studied were positive for nestin, CXCR-4, CD44, and c-kit. Immunohistochemistry identified cells positive for CD133, Pax6, Musashi, nestin, Sox2, ABCB5, and CD68 predominantly at the invading tumor front. C-terminal primers interacting with CD133 splice variant s2 detected a novel variant lacking exon 27. Differential expression of CD133 splice variants was found in iris, ciliary body, retina, and retinal pigment epithelium/choroid as well as in uveal melanoma cell lines. mRNA for nestin, Sox2, and Musashi was present in all studied cell lines. Uveal melanoma such as cutaneous melanoma may therefore contain CSCs. Further experiments are needed to isolate stem cell marker-positive cells, to evaluate their functional properties and to explore therapeutical approaches to these putative CSCs in uveal melanoma. Melanoma Res 21:405-416 (C) 2011 Wolters Kluwer Health vertical bar Lippincott Williams & Wilkins. Show less
Thill, M.; Däubener, S.; Konen, W.; Bäck, T.H.W. 2019
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... Show moreReal-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. Show less