SonoSite titan, User Manual Supplement

Page: 14 / 144

8 Chapter 6: Safety
Output Display
Mechanical and Thermal Indices Output Display Accuracy
The accuracy result for the mechanical index (MI) is stated statistically. With 90% confidence, 90% of
the measured MI values will be 16% to –31% of the displayed MI value, or +0.2 of the
displayed value, whichever value is larger.
The accuracy result for the thermal index (TI) is stated statistically. With 90% confidence, 90% of the
measured TI values will be 26% to –50% of the displayed TI value, or +0.2 of the displayed
value, whichever value is larger. The values equate to +1dB to –3dB.
A displayed value of 0.0 for MI or TI means that the calculated estimate for the index is less than 0.05.
Factors that Contribute to Display Uncertainty
The net uncertainty of the displayed indices is derived by combining the quantified uncertainty from
three sources; measurement uncertainty, system and transducer variability, and engineering
assumptions and approximations made when calculating the display values.
Measurement errors of the acoustic parameters when taking the reference data are the major source
of error that contributes to the display uncertainty. The measurement error is described in the section
on Acoustic Measurement Precision and Uncertainty included in the system user guide.
The displayed MI and TI values are based on calculations that use a set of acoustic output
measurements that were made using a single reference ultrasound system with a single reference
transducer that is representative of the population of transducers of that type. The reference system
and transducer are chosen from a sample population of systems and transducers taken from early
production units, and they are selected based on having an acoustic output that is representative of
the nominal expected acoustic output for all transducer/system combinations that might occur. Of
course every transducer/system combination has its own unique characteristic acoustic output, and
will not match the nominal output on which the display estimates are based. This variability between
systems and transducers introduces an error into displayed value. By doing acoustic output sampling
testing during production, the amount of error introduced by the variability is bounded. The
sampling testing ensures that the acoustic output of transducers and systems being manufactured
stays within a specified range of the nominal acoustic output.
Another source of error arises from the assumptions and approximations that are made when
deriving the estimates for the display indices. Chief among these assumptions is that the acoustic
output, and thus the derived display indices, are linearly correlated with the transmit drive voltage of
the transducer. Generally, this assumption is very good, but it is not exact, and thus some error in the
display can be attributed to the assumption of voltage linearity.
Related Guidance Documents
• Information for Manufacturers Seeking Marketing Clearance of Diagnostic Ultrasound Systems
and Transducers, FDA, 1997.
• Medical Ultrasound Safety, American Institute of Ultrasound in Medicine (AIUM), 1994. (A copy is
included with each system.)
• Acoustic Output Measurement Standard for Diagnostic Ultrasound Equipment, NEMA UD2-2004.