Biomedical Engineering Reference
In-Depth Information
junction boxes with excess wiring hanging in circular
loops contributed to the radiation of the pulses. Properly
closing the junction boxes and replacing the strobe lights
with lower intensity devices within fire code guidelines
resolved the situation.
and the leads laid out, unterminated, on the couch on
which patients were placed. During this procedure the
clinical engineer observed that, when pressure was ap-
plied to the couch cushion, the baseline of the EMG
machine would vary synchronously. Several additional
adjustments to the sensitivity (gain) of the device and the
time base were made. It then appeared that any motion
in the immediate vicinity of the device would cause this
baseline shift. Based on the results of these tests and of
an investigation that determined the material composi-
tion of the environment (vinyl cushions on the couch and
highly waxed vinyl floor), the engineer concluded that
the cause of the interference with the device was elec-
trostatic, not electromagnetic. Furthermore, the in-
termittent appearance of the problem was attributed to
the fact that hydrotherapeutic baths were located two
doors away from the EMG room. Their intermittent
operation would raise the humidity sufficiently to reduce
the potential for the generation of intense electrostatic
charges. This was a decidedly different ghost hunt.
Risk prevention
The following is an example of how the application
of footprinting and fingerprinting may have prevented
an EMI incident. The diagnostic imaging department
bought new telemetry equipment. Footprinting revealed
that the level of incidental emissions in that department
was approximately 67 microvolts. Fingerprinting a rep-
resentative telemetry transmitter showed that the
transmitted energy level at the standard 1-meter distance
was 12 microvolts above the background level.
The defined risk was that there was not sufficient
signal-to-noise ratio to preclude intolerably long periods
of loss of useable signal throughput. Testing the telemetry
transmitter (fingerprinting) is done under controlled
conditions at a fixed distance that can not be maintained
under real world conditions. A patient wearing a teleme-
try transmitter cannot be expected to maintain a 1-meter
distance from the receiving antennas. As the distance
between a patient and the receiving antennas increases,
the received signal degrades because of the various to-
pographical conditions and the inverse square law. As the
signal degrades, the level of acceptable data degrades
accordingly. If the degradation is significant, there is
a chance that the telemetry system will not be able to
recognize the emergency if a patient is in cardiac distress.
Interference not caused by EMI
One of the recent trends related to EMI is that of
a manufacturer's technical problems being attributed to
EMI. Although EMI may play a significant role in the
performance degradation in certain clinical devices, it is
not always the cause. One example involved radiological
equipment. X-ray films from one of the radiology labs
displayed an artifact described by the radiologists as
''chicken scratches.'' This artifact was present primarily
during a Temporo-Mandibular Joint (TMJ) procedure.
This artifact was initially attributed to either conducted
or radiated EMI, specifically conducted EMI from the
power lines. An outside consultant measured radiation
from the power lines, but there was insufficient indication
that the artifact was due to the power system. After sev-
eral months, the biomedical engineering department was
called in for consultation.
Initial investigation, including timing the artifacts as
they appeared on the film, indicated some degree of
synchronicity (timing repeatability). This implied that
the interference was time-locked (synchronized) within
the radiology system. Due to the characteristics of the
interference, such as the timing (frequency) of the in-
terference in relationship to the scan rate, a probe was
designed, constructed, and tuned to resonate within
the frequency range of the suspected interfering signal.
A thorough investigation of the areas surrounding the
radiology suite included probing of the three-phase
electrical service panels that provided the distribution
of power to the suite. No high levels of radiated elec-
tromagnetic energy were encountered within the fre-
quency of interest. The investigation was then conducted
Interference of another type
But not all ghosts are due to radiated or conducted
electromagnetic fields. For example, a report was re-
ceived of intermittent interference to an EMG device in
the physical therapy department of the hospital. The
department felt that the culprit was the MRI system
located directly above the area containing the victim
device. During footprinting, measurements entailing
more than the normal broad spectrum procedures were
indicated. In this application, the fingerprinting equip-
ment was set up to measuring any radiated electromag-
netic fields at the resonant frequency of the MRI that
might leak from the shielded room environment that
contains the MRI system. During a period of several
hours, no leakage was detected. A broader spectrum scan
showed that this department was in a remarkably quiet
location with respect to RF energy.
Further interviews with the doctors and staff led the
clinical engineer to perform a somewhat unorthodox
series of tests. The filtration on the EMG was broadened
