Source:Academic Radiology
Author(s): Wenli Cai, Seth M. Steinberg, Miriam A. Bredella, Gina Basinsky, Bhanusupriya Somarouthu, Scott R. Plotkin, Jeffrey Solomon, Brigitte C. Widemann, Gordon J. Harris, Eva Dombi
ObjectivesPlexiform neurofibromas (PNs) are complex, histologically benign peripheral nerve sheath tumors that are challenging to measure by simple line measurements. Computer-aided volumetric segmentation of PN has become the recommended method to assess response in clinical trials directed at PN. Different methods for volumetric analysis of PN have been developed. The goal of this study is to test the level of agreement in volume measurements and in interval changes using two separate methods of volumetric magnetic resonance imaging analysis.MethodsThree independent volume measurements were performed on 15 PN imaged at three time-points using 3DQI software at Massachusetts General Hospital (MGH) and National Cancer Institute (NCI) and MEDx software at NCI.ResultsMedian volume differences at each time-point comparing MGH-3DQI and NCI-3DQI were −0.5, −4.2, and −19.9 mL; comparing NCI-3DQI and NCI-MEDx were −21.0, −47.0, and −21.0 mL; comparing MGH-3DQI and NCI-MEDx were −10.0, −70.3, and −29.9 mL. Median differences in percentage change over time comparing MGH-3DQI and NCI-3DQI were −1.7, 1.1, and −1.0%; comparing NCI-3DQI and NCI-MEDx were −2.3, 3.3, and −1.1%; comparing MGH-3DQI and NCI-MEDx were −0.4, 2.0, and −1.5%. Volume differences were <20% of the mean of the two measurements in 117 of 135 comparisons (86.7%). Difference in interval change was <20% in 120 of the 135 comparisons (88.9%), while disease status classification was concordant in 115 of 135 comparisons (85.2%).ConclusionsThe volumes, interval changes, and progression status classifications were in good agreement. The comparison of two volumetric analysis methods suggests no systematic differences in tumor assessment. A prospective comparison of the two methods is planned.
http://ift.tt/2zZ6SnK
Δεν υπάρχουν σχόλια:
Δημοσίευση σχολίου