The Design and Use of the ICRU/AAPM CT Radiation Dosimetry Phantom: An Implementation of AAPM Report 111-6BD3C7AC

Although the CTDI phantom and its associated pencil ionization chamber measurement methodology have proven to be very useful, they suffer from limitations that have been critically examined over the past several years. As a consequence, Task Group 111 was formed by the AAPM to address some of the concerns that had been raised. Among these concerns are the following:

1. CTDI100 is a surrogate for the dose at the center of a scan of a single, fixed length, 100 mm, and excludes the dose that would accumulate for longer scans. “This underestimation is systematic, applying to narrow and wide beams alike, and slowly becomes larger with increasing width of the z-axis collimation.”¹
2. Some scanners employ very wide beams, approaching or even wider than the length of the 100 mm pencil chamber, rendering the CTDI paradigm completely unsuitable for characterizing the dose.
3. CTDI may be inappropriate for stationary table applications, particularly for beams wider than 100 mm. AAPM Report 111¹ offered several recommendations for a new measurement methodology and suggested several phantom designs.

Task Group 200 and the ICRU’s Committee on Radiation Dose and Image-Quality Assessment in Computed Tomography have jointly developed a phantom design and robust measurement schemes that follow the methodology of AAPM Report 111 and are suitable for a wide range of CT scanner designs and scanning conditions. Several prototype phantoms were built by a research group at the University of California–Davis (UC Davis), and these phantoms have been tested at several centers around the United States as well as in England.
The purpose of this current report is to (a) describe the design of the phantom and (b) suggest a broadly applicable measurement methodology that overcomes the limitations of CTDI100 and the met- rics derived from it, such as CTDI_vol. The resulting measurement procedures have been developed for conventional MDCT scanners, including models with wide (16 cm, for example) beams. However, the application of this methodology to flat panel and specialized cone-beam CT systems presents special challenges that are briefly discussed in Appendix 1. The solutions to these particular problems are beyond the scope of this report.