Biomedical Engineering Reference
In-Depth Information
the patients' data examined in time can be recorded in a database linked to the CAD
system for storage or retrieval.
One important aspect for guaranteeing the quality of follow-up results is the
reproducibility
. For instance, when a reconstruction that has been created some time
ago is to be compared to a recent one, the same parameters for the visualization have
to be used in order to guarantee that obvious changes to the symptoms are not due
to the changed nature of the visualization. For the sake of reproducibility and other
legal issues, a CAD should be able to save the entire “session”, which represents all
the actions a user took to generate all the patients' data (e.g. anatomical elements,
clinical data, diagnostic parameters, annotations). This includes the current state of
work along with the history of all changes, like the results of certain computed para-
meters. By saving this information in the patients' database, CAD systems provide
features useful for quantitative and qualitative comparisons between exams acquired
at different times.
12.3.2.1 Quantitative Evaluation
Follow-up studies during the treatment involve the acquisition of images at various
stages, such as three, six, and twelve months after the beginning of the therapy. The
evaluation of these images involves a comparison, for example, with respect to tumor
growth or to other clinical indicators (e.g. erosion scoring for RA, blood sugar for
diabetes).
For example, the difference between numerical results of clinical parameters
obtained using one or more of the techniques described in the previous sections
or using graphical elements, such as plots or charts, leads to the quantitative evalua-
tion of the disease regression or progression. This helps the physician to eventually
modify the treatment plan.
12.3.2.2 Qualitative Evaluation
The visualization of anatomical elements in 2D or 3D images through overlapping
or side-by-side views leads to qualitative evaluation. For radiologists, some diseases
have become easier to identify, grade and follow with the help of 3D visualization.
For instance, typical 3D techniques include the synchronized rotation of the whole
dataset, synchronized zooming towards relevant structures, synchronized movement
of clipping planes, synchronized movement of the camera along a certain path (for
example, through tubular structures), and synchronized gradual changes of the trans-
parencies of objects.
