Mining of Sensor Data in Healthcare: A Survey

Abstract

Historically, healthcare has been mainly provided in a reactive manner that limits its usefulness. With progress in sensor technologies, the instrumentation of the world has offered unique opportunities to better observe patients physiological signals in order to provide healthcare in a more proactive manner. To reach this goal, it is essential to be able to analyze patient data and turn it into actionable information using data mining. This chapter surveys existing applications of sensor data mining technologies in healthcare. It starts with a description of healthcare data mining challenges before presenting an overview of applications of data mining in both clinical and non clinical settings.

Computational Health Informatics in the Big Data Age

“…Data missing for an intentional reason or because of irrelevancy to a current clinical problem is considered nonrecoverable and thus deleted [Fialho et al 2012]. -Data synchronization: Sensor data is reported at different rates with timestamps based on their internal clocks [Sow et al 2013]. However, the clocks across sensors are often not synchronized.…”

“…-Data cleaning: This step involves the removal of noise in healthcare data, which includes artifacts [Singh et al 2011;Mao et al 2012] and frequency noise in clinical data [Sow et al 2013]. For example, thresholding methods are used to remove incompliant measurements [Fauci et al 2008;Apiletti et al 2009] and low-pass/high-pass filtering tools are usually applied to remove frequency noise in sensor signals [Hu et al 2008;Frantzidis et al 2010;].…”

“…It is necessary to make reasonable assumptions and derive alignment strategies to synchronize sensor data [Jané et al 1991;Martinez Orellana et al 2013]. -Data normalization: This step is often required to cope with differences in the data recording process [Sow et al 2013]. For example, a daily heart rate may represent a daily average heart rate or a measurement during a specific time range.…”

“…A typical healthcare data preprocessing procedure usually includes the following steps depending on the source and format of the data [Banaee et al 2013;Sow et al 2013]:…”

Processing and Adding Vibrancy to Sensor Data

“…These data sets will likely continue to grow unabated. However, such data sets are vastly under-utilized, despite their potential to deliver insights into the future well-being of patients, particularly for time-critical scenarios (Sow et al, 2013). And various sensor types are already being used to track the environment.…”