| Summary: | The oscillometric method for blood pressure measurement has been known
for over a century. It was overshadowed by the classic stethoscope and
cuff method until more recently when its ease of automation became
useful. It is now found in the automated blood pressure cuffs
used in hospitals, doctor's offices, pharmacies, and devices sold
for home use. It still challenges accurate blood pressure measurement,
however, due to its difficulty in compensating for pregnancy, age,
hypo-, and hypertension.
Global sensitivity analysis methods were used to develop a model that
focuses on the most important system parameters. The most influential
biological and design parameters were identified allowing the removal
or fixing of less influential parameters, and the replacement of
subsystems with linear models, with minimal effect on the
overall system accuracy.
The developed model allows for the investigation and development of
new methods for extracting parameter information from the
oscillometric signals. This is illustrated by the development of a
method to extract the artery's cross-sectional area from standard
oscillometric output.
The system design requirements for accurate measurement of blood
pressure are examined and discussed with recommendations for system
parameter adjustment. The model's performance and usefulness is
highlighted with modelled case studies of potential real-world
applications for subjects with parameters or inputs for which the
oscillometric method would find it difficult to compensate. Through
the use of the developed model to compensate for the system errors,
the measurement error can be reduced by half.
Highlighting the important system parameters allows the engineer to
focus on choosing the design parameters over which she or he has
control. The system model provides the ability to experiment with the
cuff design choices and provides information regarding system
performance under conditions that are historically difficult to
measure accurately. The developed model's usefulness is illustrated by
applying it to parameter extraction, and to the compensation of
oscillometric blood pressure readings.
|
|---|
