Filtern
Dokumenttyp
- Wissenschaftlicher Artikel (Fachzeitschriften) (7) (entfernen)
Sprache
- Englisch (7)
Volltext vorhanden
- ja (7)
Gehört zur Bibliographie
- nein (7)
Schlagworte
- Maschinelles Lernen (4)
- Anomalie <Medizin> (2)
- Anomalieerkennung (2)
- Convolutional Neural Network (2)
- anomaly detection (2)
- machine learning (2)
- Algorithmus (1)
- Apoptosis (1)
- Bilderkennung (1)
- Bioklima (1)
- Boden (1)
- Bodenorganismus (1)
- Bodentemperatur (1)
- CNN (1)
- CO2 emissions (1)
- DNN (1)
- Deep learning (1)
- Diskriminanzanalyse (1)
- EHR (1)
- Elektronische Patientenakte (1)
- Gesundheitsökonomie (1)
- Herzmuskelkrankheit (1)
- ICU (1)
- Intensivpflege (1)
- Intensivstation (1)
- Klimaschutz (1)
- Klimaänderung (1)
- Kohlendioxidemission (1)
- Künstliche Beatmung (1)
- Künstliche Intelligenz (1)
- Laborparameter (1)
- Lufttemperatur (1)
- Medizinische Informatik (1)
- Messung (1)
- Meteorologische Station (1)
- Mikroklima (1)
- Neuronales Netz (1)
- Personenkraftwagen (1)
- Petri nets (1)
- Rekursives neuronales Netz (1)
- Selbstorganisierende Karte (1)
- Sepsis (1)
- Statistik (1)
- VentAI (1)
- apoptosis (1)
- artificial intelligence (1)
- bioclimatic variables (1)
- classification (1)
- climate policy (1)
- convolutional neural networks (1)
- critical care (1)
- critically-ill patient (1)
- deep learning (1)
- efficiency trade-offs (1)
- global maps (1)
- image recognition (1)
- intensive care (1)
- lab values (1)
- lightGBM (1)
- linear discriminant analysis (LDA) (1)
- medical consumables (1)
- medical economics (1)
- medical informatics (1)
- microclimate (1)
- modeling (1)
- near-surface temperatures (1)
- necroptosis (1)
- optimization (1)
- passenger cars (1)
- reinforcement learning algorithm (1)
- self organizing maps (SOM) (1)
- septic cardiomyopathy (1)
- small data (1)
- soil temperature (1)
- soil-dwelling organisms (1)
- temperature offset (1)
- time series classification (1)
- vehicle mass (1)
- vehicle power (1)
- ventilation (1)
- weather stations (1)
Institut
Numerous research methods have been developed to detect anomalies in the areas of security and risk analysis. In healthcare, there are numerous use cases where anomaly detection is relevant. For example, early detection of sepsis is one such use case. Early treatment of sepsis is cost effective and reduces the number of hospital days of patients in the ICU. There is no single procedure that is sufficient for sepsis diagnosis, and combinations of approaches are needed. Detecting anomalies in patient time series data could help speed the development of some decisions. However, our algorithm must be viewed as complementary to other approaches based on laboratory values and physician judgments. The focus of this work is to develop a hybrid method for detecting anomalies that occur, for example, in multidimensional medical signals, sensor signals, or other time series in business and nature. The novelty of our approach lies in the extension and combination of existing approaches: Statistics, Self Organizing Maps and Linear Discriminant Analysis in a unique and unprecedented way with the goal of identifying different types of anomalies in real-time measurement data and defining the point where the anomaly occurs. The proposed algorithm not only has the full potential to detect anomalies, but also to find real points where an anomaly starts.