Description
Critically ill patients receive acute, complex and dynamic care within the intensive care unit (ICU), where they are monitored by a variety of biosensors that generate continuous data [1], [2]. Those sensors, such as the electrocardiogram (ECG), for instance, are sampled at hundreds of hertz (Hz). The capture and analysis of this data has been a challenge, due to the limited availability of methods and software to enable seamless capture, processing and analysis. While some open-source approaches exist for selected medical devices vendors [3], the availability of such software to perform preprocessing functions, including identifying noisy segments of the data and integrating clinical patient binding from hospital systems have been limited. We introduce an approach to support the preprocessing of real-time physiological data and stage such data in a NoSQL environment to enable scalable and efficient preprocessing.
Describe the new knowledge and additional skills the participant will gain after attending your presentation.: Data pre-pocessing techniques, Match-merging of collections/datasets on multiple conditions in MongoDB.
Authors:
Lokesh Chinthala (Presenter)
University of Tennessee Health Science Center
Akram Mohammed, University of Tennessee Health Science Center
Rishikesan Kamaleswaran, University of Tennessee Health Science Center
Presentation Materials: