Evaluation of Semiautomated Measurements of Mesothelioma Tumor Thickness on CT Scans1

Modeling of mesothelioma growth demonstrates weaknesses of current response criteria

“…Pass et al [18] demonstrated a relationship between mesothelioma tumor volume and survival; however, the routine radiologic assessment of tumor volume through manual delineation of the complete tumor margin on CT is impractical. We are currently investigating computerized techniques for the measurement of mesothelioma extent in CT scans [16,19], with the ultimate objective of automatically capturing tumor volume as the definitive method for accurately assessing tumor response to therapy [20]. The present study demonstrates the conceptual shortcomings of the current tumor response criteria, an initial step toward undertaking the exploration of a more definitive response endpoint.…”

Current state and future directions of pleural mesothelioma imaging

“…This study also found statistically significant differences in 8 out of 10 pairwise comparisons of the measurements obtained by five observers. Automated and semi-automated measurement techniques are being developed in response to these concerns [15][16][17]. In an observer evaluation study, 75% of the tumor thickness measurements derived from a semi-automated method were accepted without modification by observers, and 80% of the altered measurements were changed by less than 2 mm [16].…”

Interactive Modeling and Evaluation of Tumor Growth

“…Zacharaki et al 5 applied a multimodal registration and fusion approach to simulate 3D tumor growth, but results are reported only for registered images with no significant temporal gap. A hybrid approach for semi-automated measurement of lung tumor thickness was proposed by Armato et al 6 In their approach, the user indicates initial endpoints along the outer tumor margin; these user-defined endpoints are refined by further processing, and an estimate of the maximum tumor diameter is obtained. However, the maximum tumor diameter is a limited measure to quantify the tumor growth, since the shape of a tumor can be complex and contain concavities.…”