It is increasingly recognised that understanding the condition of a patient may require more extensive monitoring than has happened in the past.  A cardiovascular problem may be intermittent and may not be readily apparent during the short monitoring time at the clinic.  Blood pressure measurements taken at the clinic may not accurately record the state of the patient in a more normal environment.  Chronic conditions in particular need frequent monitoring yet need to be balanced by the high costs of bringing a patient to a hospital.

The barriers to remote, patient monitoring have dropped as small, cheap sensors are now available to cover a variety of physiological and chemical measurements.  However, large scale introduction of these measurement technologies will create a problem for health systems as they struggle to redefine their patient management process in order to handle the resulting data explosion.

SDE’s pattern matching technology can provide alerting mechanism for numerous conditions reducing the communication of data flow to the clinician.   In a hospital environment, SDE’s pattern recognition is a potential application that Cybula has investigated  to identify ECoG patterns in brain injury and in identifying sleep patterns.


  • Quickly visualise large time series data sets.
  • Search for example patterns in the data- solve the ‘needle in the haystack’ problem.
  • Detect complex patterns over multiple time series.
  • Access to remote data held in distributed databases.
  • Filtering and combine data to analyse complex events
  • Detect spikes in raw data.
  • Read a large number of standard file types.

Previous applications

EEG Analysis

Typical EEG data can be large and contain many channels. SDE’s visualisation capability allows users to browse such data very easily. Data can be displayed in a single window or over multiple windows and ‘play’ the data for ease of viewing. Any events with known activity patterns can be searched for using SDE’s built in search toolset. Data can also be filtered to remove artefacts in the data using a number of in built filtering methods.

Spike Train Analysis

One of the most common tasks when looking at electrophysiology data is the identification of spikes in the raw data. SDE uses its powerful pattern match capabilities to allow users to identify spikes in their data. Users can also specify their own spike profiles and use SDE’s in built pattern search engine to find such events.