Computer assisted tomography (CAT)

Computer assisted tomography (CAT) reveals the structure of the living brain. The head is placed among a source that emits a narrow beam of X-rays and an X-ray detector. A series of measurements is made of X-ray transmission. The detector and source are rotated as a pair through a small angle and a further series of measurements taken. This is repeated until the detector and source have rotated by 180?. The radiodensity of each region of the head is computed from the transmission data for all of the beams which have traversed that region, and the results visually displayed. It can provide a view by a single slice of brain lying at a known orientation. Through moving the head at right angles to the orientation plane for a little distance another section can be imaged. This is reeated until the overall brain has been scanned.

The major element here is the algorithm and the computer software to implement it which calculates the radiodensity for each point in the brain slice; this is computerized tomography. It is an  instance of an  inverse problem that begins with  a data set from which initial parameters, in this  case  source location, must be  calculated. It contrasts with forward problems in that the source location is known and it is the data set which is calculated. The difficulty with inverse problems is they do not have unique solutions. Therefore they have to be constrained through assumptions and prior modeling based on earlier results to search the most likely solution.

CAT can differentiate tissues that differ in X-ray opacity through 1percent (the lower the density the darker the image) with a spatial resolution of about 0.5 mm. Blood vessels can be seen through injection of radio-opaque dyes.

                   Figure: Computer assisted tomography (CAT). Arrows depict the rotation of the scanner.