Abstract
In the photodynamic diagnosis of brain surgery, neurosurgeons visually verify fluorescence signals from a brain tumor. Therefore, they can detect strong fluorescence signals but not weak ones. The purpose of this research is to build a new imaging system to highlight such weak fluorescence signals during brain tumor resection.
The proposed system comprised a light source for the fluorescence excitation, a video camera to record fluorescence, and a PC for image processing.
We recorded 630-nm fluorescence from a brain tumor as a 24-bit color-depth image video and evaluated the RGB components in each pixel of the fluorescence. The R-component of this fluorescence had pixel values of 4-10 for the fluorescence and values of 1-3 for noise. The fluorescence was invisible on the recorded video and was only included in the R-component. To highlight weak fluorescence, we wrote a program to sum any capture images to improve the SNR and to emphasize the R-components of the images. Applying the image-processing program to a fluorescence video, we were able to observe the weak fluorescence on the PC monitor.
We confirmed that the proposed system that we built can highlight weak fluorescence from a brain tumor.
