Moving targets in 4D-CTs versus MIP and AIP: comparison of patients data to phantom data

Purpose: Maximum (MIP) and average intensity projection (AIP) CTs allow rapid definition of internal target volumes in a 4D-CT. The purpose of this study was to assess the accuracy of these techniques in a large patient cohort in combination with simulations on a lung phantom. Methods: 4DCT data from a self-developed 3D lung phantom and from 50 patients with lung tumors were analyzed. ITVs were contoured in maximum (ITVMIP) and average intensity projection (ITVAIP) and subsequently compared to ITVs contoured in 10 phases of a 4D-CT (ITV10). In the phantom study additionally a theoretical target volume was calculated for each motion and compared to the contoured volumes. Results: ITV10 overestimated the actual target volume by 9.5% whereas ITVMIP and ITVAIP lead to an underestimation of -1.8% and -11.4% in the phantom study. The ITVMIP (ITVAIP) was in average -10.0% (-18.7%) smaller compared to the ITV10. In the patient CTs deviations between ITV10 and MIP/AIP were significantly larger (MIP: -20.2% AIP: -33.7%) compared to this. Tumors adjacent to the chestwall, the mediastinum or the diaphragm showed lower conformity between ITV10 and ITVMIP (ITVAIP) compared to tumors solely surrounded by lung tissue. Large tumor diameters (> 3.5 cm) and large motion amplitudes (> 1 cm) were associated with lower conformity between intensity projection CTs and ITV10-. Conclusion: The application of MIP and AIP in the clinical practice should not be a standard procedure for every patient, since relevant underestimation of tumor volumes may occur. This is especially true if the tumor borders the mediastinum, the chest wall or the diaphragm and if tumors show a large motion amplitude.