Abstract
In Japan, cerebrovascular disease is the fourth cause of death and the first cause
of bedridden patients. Recently, cerebral infarction has been under serious consideration due
to the westernization of the diets and an increase in geriatric diseases. To establish a method
for image-based diagnosis of cerebral apoplexy, the authors developed a phantom that could
correctly evaluate disease detection by image-processing and that could visualize disease
using X-ray CT imaging, while evaluating the imaging conditions. Traditionally, it has been
difficult to visualize acute cerebral infarction from images produced by X-ray CT. Therefore, a
frequency-enhancement processing technique that applies an unsharp mask filtering was
developed. We acquired a high-frequency image containing many frequency bands of a
cerebral infarction region, added the acquired image to an original image, and created an
algorithm that could enhance these frequency bands and decrease image noise. We then
tested the algorithm by comparing the image from the X-ray CT process to the simulation
result obtained using a new cerebral stroke phantom developed by us. An outline of the
algorithm and the results obtained are presented.
