Determination of the weighted CT dose index in modern multi-detector CT scanners.

The aim of the present study was to (a) evaluate the underestimation in the value of the free-in-air (CTDI(air)) and the weighted CT dose index (CTDI(w)) determined with the standard 100 mm pencil chamber, i.e. the CTDI(100) concept, for the whole range of nominal radiation beam collimations selectable in a modern multi-slice CT scanner, (b) estimate the optimum length of the pencil-chamber and phantoms for accurate CTDI(w) measurements and (c) provide CTDI(w) values normalized to free-in-air CTDI for different tube-voltage, nominal radiation beam collimations and beam filtration values. The underestimation in the determination of CTDI(air) and CTDI(w) using the CTDI(100) concept was determined from measurements obtained with standard polymethyl-methacrylate (PMMA) phantoms and arrays of thermoluminescence dosimeters. The Monte Carlo N-Particle transport code was used to simulate standard CTDI measurements on a 16-slice CT scanner. The optimum pencil-chamber length for accurate determination of CTDI(w) was estimated as the minimum chamber length for which a further increase in length does not alter the value of the CTDI. CTDI(w)/CTDI(air) ratios were determined using Monte Carlo simulation and the optimum detector length for all selectable tube-voltage values and for three different values of beam filtration. To verify the Monte Carlo results, measured values of CTDI(w)/CTDI(air) ratios using the standard 100 mm pencil ionization chamber were compared with corresponding values calculated with Monte Carlo experiments. The underestimation in the determination of CTDI(air) using the 100 mm pencil chamber was less than 1% for all beam collimations. The underestimation in CTDI(w) was 15% and 27% for head and body phantoms, respectively. The optimum detector length for accurate CTDI(w) measurements was found to be 50 cm for the beam collimations commonly employed in modern multi-detector (MD) CT scanners. The ratio of CTDI(w)/CTDI(air) determined using the optimum detector length was found to be independent of beam collimation. Percentage differences between measured and calculated corresponding CTDI(w)/CTDI(air) ratios were always less than 8% for head and less than 5% for body PMMA phantoms. In conclusion, the CTDI(air) of MDCT scanners may be measured accurately with a 100 mm pencil chamber. However, the CTDI(100) concept was found to be inadequate for accurate CTDI(w) determination for the wide beam collimations commonly used in MDCT scanners. Accurate CTDI(w) determination presupposes the use of a pencil chamber and PMMA phantoms at least 50 cm long.