Pattern Classification Methods for Analysis and Visualization of Brain Perfusion CT Maps - Computational Intelligence Paradigms in Advanced Pattern Classification - page 149

with cell death in many minutes to hours [18]. CBF of less than 10 ml/100 g of tis-
sue per minute cannot be tolerated beyond a few minutes before infarction occurs
causing permanent brain cells damage [1], [12], [18].
Because of some factors [13] the true CBF and CBV values in individual cases
may be underestimated in a manner, which is difficult to predict. As the result of
that some authors prefers to use relative values of CBF and CBF (appropriately
rCBF and rCBV). Authors in [13] states that a relative comparison of cerebral
blood flow within corresponding areas of both hemispheres of the brain is possible
without any limitations because the error of measurement is the same for both
high and low CBF values (Table 2).
Table 2 Average relative values of perfusion parameters (CBF and CBV) and their
interpretation
rCBF decrease RCBV Interpretation
0.60 - 0.70 Tissues can be salvaged [30]
0.20 - 0.30 Tissues will eventually become infarcted [30]
0.62 +- 0.17 0.78 +- 0.18 Tissues can be salvaged [30]
0.34 +- 0.20 0.43 +- 0.22 Tissues will eventually become infarcted [16]
0.48 0.60 Threshold for distinguish infarcted regions
(discriminant analysis) [16]
With PWI (perfusion - weighted images) it is possible to detect the perfusion
deficit area by CTP (computed tomography perfusion) before noncontrast CT re-
veals early ischemic changes [30]. The size of the perfusion deficit determined by
CTP correlates with the prognosis of patients. In [1] authors state that compared
with diffusion-weighted MR imaging (DWI), the extent of abnormality seen in ini-
tial CBV images is a better overall predictor of final infarct size. Also according to
[19] CBV should reflect the approximate minimal final infarct size of brain tissues.
Both CBF and CBV have prognostic values in evaluation of ischemic evolu-
tion. In many cases simultaneous analysis of both CBF and CBV perfusion para-
meters enables accurate analysis of ischemia visualized brain tissues and predict
its further changes permitting a not only a qualitative (like CT angiography does)
but also quantitative evaluation of the degree of severity of the perfusion distur-
bance which results from the particular type of occlusion and collateral blood.
According to [12], [30] an increase in CBV reflects the collateral pathway or
auto regulation and a decrease in CBV indicate an unfavorable state. A decrease in
CBF is a highly sensitive as well as specific finding in predicting an infarction.
In acute ischemic stroke, a collapse of vessels at low CBV is likely to occur on-
ly after prolonged, severe reductions in CBF, and continued collapse would be