Constructing a Bio-Signal Repository from an Intensive Care Unit for Effective Big-data Analysis: Poster Abstract

Analyzing large quantities of bio-signal data can lead to new findings in patient status diagnosis and medical emergency event prediction. Specifically, improvements in machine learning schemes suggest that by inputting clinical waveforms, designing mechanisms to predict medical emergencies, such as ventricular arrhythmia or sepsis, can soon be possible. However, we are still lacking the data-vaults that provide such clinically useful bio-signal data. With the goal of providing such an environment, this work focuses on developing a data repository for bio-signals collected from a hospital's intensive care init (ICU). Specifically, we design our data collection system to effectively store data from at-bed patient monitors and also integrate sensing information from bed-embedded sensing platforms, which allow filtering of noisy bio-signal samples caused by motion artifacts.