Consensus guide on CT-based prediction of stopping-power ratio using a Hounsfield look-up table for proton therapy

Abstract

Background and purpose: Studies have shown large variations in stopping-power ratio (SPR) prediction from computed tomography (CT) across European proton centres. To standardise this process, a step-by-step guide on specifying a Hounsfield look-up table (HLUT) is presented here.

Materials and methods: The HLUT specification process is divided into six steps: Phantom setup, CT acquisition, CT number extraction, SPR determination, HLUT specification, and HLUT validation. Appropriate CT phantoms have a head- and body-sized part, with tissue-equivalent inserts in regard to X-ray and proton interactions. CT numbers are extracted from a region-of-interest covering the inner 70% of each insert in-plane and several axial CT slices in scan direction. For optimal HLUT specification, the SPR of phantom inserts is measured in a proton beam and the SPR of tabulated human tissues is computed stoichiometrically at 100 MeV. Including both phantom inserts and tabulated human tissues increases HLUT stability. Piecewise linear regressions are performed between CT numbers and SPRs for four tissue groups (lung, adipose, soft tissue, and bone) and then connected with straight lines. Finally, a thorough but simple validation is performed.

Results: The best practices and individual challenges are explained comprehensively for each step. A well-defined strategy for specifying the connection points between the individual line segments of the HLUT is presented. The guide was tested exemplarily on three CT scanners from different vendors, proving its feasibility.

Conclusion: The presented step-by-step guide for CT-based HLUT specification with recommendations and examples can contribute to reduce inter-centre variations in SPR prediction.

Keywords: Hounsfield look-up table; Proton range prediction; Proton therapy; Single-energy CT; Stoichiometric calibration; Stopping-power ratio.

Figure 1:

(A) Individual datapoints and regression lines for the different tissue groups. (B) Generated Hounsfield look-up table (HLUT) with the regression lines limited to the tissue group-specific CT number interval and connection lines in-between. The two insets show the connection between neighbouring tissue groups. The slope is positive in each curve segment.

Figure 2:

(Left) Calibration datapoints (phantom inserts and tabulated human tissues) and the resulting HLUTs for the head and body phantom for three different CT scanners. (Right) SPR deviations of the calibration tissues in the head and body phantom from the respective HLUTs as well as from the HLUT for the averaged CT numbers, summarised in the respective tissue groups. For the boxplots, the boxes cover the interquartile range (25th to 75th percentile), the whiskers extend to 1.5 times the interquartile range, and outliers beyond this range are marked with a ‘+’ symbol. The median and mean are indicated by the horizontal line and square within the box, respectively. Due to the limited number of datapoints, lung tissues are plotted individually.