Learning Center
Plans & pricing Sign in
Sign Out
Get this document free

ECG Signal Quality Measurement


									ECG Signal Quality Measurement

    Client: Alan Clapp - Senior Electrical Engineer, GE Medical Systems
                     Advisor: John G. Webster, Ph. D
               Group Members: Paul Anheier, Michael Piché,
                        John Puccinelli, Scott Wiese
              Problem Statement:
• Although modern ECGs sufficiently
  eliminate many types of interference, more
  optimization is possible and necessary. The
  focus of the project, therefore, is to further
  improve signal quality and develop a
  reliable alert system to detect signal
  degradation whether through procedural
  guidelines and/or hardware modifications.
• Poor ECG signals can have many causes:
   – Electrical interference from other instruments/power
   – Improper electrode placement
   – Poor electrode adhesion
   – Electrode aging/degradation
• Most common causes occur at the skin/electrode
• Poor contact or old electrodes result in high
  impedance at the skin/electrode interface
• This results in a degradation of signal amplitude and
  an increased susceptibility to motion artifact

• What is deemed “good” signal quality
  is highly subjective
• It is impractical to determine a
  universally acceptable signal quality
  due to subjectivity
• Solution?...
        Proposed Solution
• Include as a feature on future
  electrocardiographs a graphical display
  of measured skin impedance over time
• Graph would have fixed scale to make
  interpretation easier
• Clinicians could make their own
  decisions on signal quality based on
  trends in the graph
Might look something like this…
 Requirements for Implementation
• We must determine a suitable scale to use
  for graph of skin resistance
• Determination of the best carrier signal to
  measure impedance (DC, ~.2 Hz, 250 Hz)
• Determine impedance level above which
  problems frequently occur
• Determination a typical response of skin
  impedance over a long time interval (24
       Carrier Signal Testing
• Goal: Identify most reliable/accurate
  carrier signal for measuring impedance.
• Skin is not perfect resistor, must determine
  behavior at different frequencies
• Candidates (at request of GE engineers)
  – DC
  – 250 Hz
  – .2 Hz
    Carrier Signal Testing
Frequency        Pros               Cons

  DC        Easy to measure    electrodes over

                               Lead fail alarm
            May not polarize
 0.2 Hz     electrodes
                               samples every 5
                               s (too long)

                               High frequency –
250 Hz      NO polarization    skin may behave
        Carrier Signal Testing
• Human Subjects Committee has
  conditionally approved public participants.
  – With this approval, we can maximize test
    subject diversity
     • ↑ subjects = ↑ skin types = more realistic results
• Considerations for Analysis
  – Input impedance of oscilloscope
  – Skin resistance changes over time
     Skin Impedance Testing
• Goal: Collect data over 24 hours of
  impedance change at skin-resistor
  interface. Use to establish a scale.
• Study is to include multiple types of
• Necessary for implementation of a
  graphical display of skin impedance
  change versus time.
          Future Directions
• Complete carrier signal testing and
• Establish resistance scale.
• Develop layout of graphical display.
• Characterize quality/response of various
• Submit to GE for review and possible
• Publish the findings.
• Graphical representation of skin
  impedance will provide useful data and
  help make decisions on signal/electrode
• Determining a signal carrier is a key
  component of representing resistance

To top