Step 3: Processing the Images

The first step is to add the sound that we recorded during image acquisition to the images. This is done using some clever program written in Matlab. The next movie shows the beginning of the song with audio recording. During the first few seconds you can hear the scanner noise. The noise is then cancelled and we count down and start the music for James to sing:

As you can imagine, singing on your back, with your head tilted backwards is not the easiest thing to do and that explains why SiVU does not sound as great as he usually does! The rest of the processing was done using OsiriX, an open source medical image software. First, we created movies of James singing (without sound) and "fly-through" movies of his brain (like on the previous page) using a selection of color scales instead of standard grey scale:


Finally, we created different surface and volume renderings of James' head:

The depressions around SiVU's ears are caused by the large headphones pushing against his head.

You can see more renderings in the final video (next page)