Biomedical Engineering Reference
In-Depth Information
a universal portal for disseminating image-intensive news and entertainment, joined recently by the
multimedia-rich Web, is a testament to our ability to immediately evaluate graphical content without
conscious, focused mental processing.
Although there are exceptions, it's often difficult for even highly trained professionals to intuitively
evaluate strings of text or tables of data so that they can act on them quickly. This is especially true
when we are inundated throughout the day with data from a variety of sources, each source
competing for attention. Everyone from aircraft pilots, drivers, anesthesiologists, and nuclear power
plant operators to molecular biologists rely on graphical displays to operate equipment and
communicate findings to others.
Consider that the typical physician understands or is at least familiar with the concepts of statistical
sensitivity and specificity as applied to the interpretation of routine laboratory test results. However,
when asked to apply these concepts to their everyday practice of reviewing tables of numerical
laboratory test values, most cannot calculate when a test result is far enough from normal to warrant
further investigation. For this reason, many laboratories report laboratory results to physicians in a
tabular, numeric form in which each value is accompanied by a normal range and a simple graphic to
show how it relates to what is generally accepted as the normal range.
The list of blood values for a male patient shown in
Figure 5-1
is representative of how simple
graphics are used at many hospitals and clinics to allow physicians to quickly visualize significantly
abnormal results. In this example, the fasting blood glucose and hematocrit levels are outside of their
normal ranges. The degree of abnormality can be calculated by looking at the range of normal
values. However, because of the difference in ranges, it isn't immediately clear that the fasting blood
glucose level is significantly out of normal range and that the hematocrit is just outside of normal.
The advantage of the graphic is that the data ranges are normalized so that the distance outside of
the normal range brackets has the same relative significance across all laboratory results.
Figure 5-1. Visualization Aids to Tabular Clinical Laboratory Data.
