Neonates who are critically ill are cared for in a neonatal intensive care unit (NICU) for continuous monitoring of their conditions. Physiological parameters such as heart rate, respiratory wave form, blood oxygen saturation, and body temperature are constantly monitored in the NICU. However, NICUs are not always equipped with a computer system for analyzing such data, identifying critical events, and providing decision support for a neonatologist. Therefore, a specialized computer system, commonly known as a data mart, should be developed for the NICU. An architectural framework for the design and development of an automated system for data collection, storage, and analysis for the NICU is proposed in this paper. Our study also deals with the implementation of advanced transformation functions such as fuzzy grouping and fuzzy lookup for data preparation and preprocessing. Furthermore, based on Kimball’s dimensional modeling, a data gathering and accumulation system, fact constellation schema (galaxy schema) has been built with previously identified neonatal processes. Finally, an information delivery component has been proposed, wherein the neonatal data can be both analyzed with different data mining algorithms and visualized with various metrics. Our pioneering work demonstrates methods that streamline the process of data collection, data storage, analysis, and decision-making, which in turn increases efficiency in the NICU and saves lives. Our study, presented in this paper, describes the design and development process and discusses its utility in the NICU. Our results indicate that data mart is best suited for effective decision-making in the NICU.
|Number of pages||23|
|Journal||Critical Reviews in Biomedical Engineering|
|Publication status||Published - 01-01-2018|
All Science Journal Classification (ASJC) codes
- Biomedical Engineering