Comparing Image Segmentation Techniques for Determining 3D Orbital Cavernous Hemangioma Size on MRI

Abstract

Purpose: To measure orbital cavernous hemangioma size using 3 segmentation methods requiring different degrees of subjective judgment, and to evaluate interobserver agreement using these methods. Methods: Fourteen patients with orbital cavernous hemangiomas were included in the study. Pretreatment T2-weighted MRIs were analyzed by 2 observers using 3 methods, including 1 user-dependent image segmentation method that required high degrees of subjective judgment (ellipsoid) and 2 parameter-dependent methods that required low degree of subjective judgment (GrowCut and k-means clustering segmentation). Interobserver agreement was assessed using Lin’s concordance correlation coefficients. Results: Using the ellipsoid method, the average tumor sizes calculated by the 2 observers were 1.68 ml (standard deviation [SD] 1.45 ml) and 1.48 ml (SD 1.19 ml). Using the GrowCut method, the average tumor sizes calculated by the 2 observers were 3.00 ml (SD 2.46 ml) and 6.34 ml (SD 3.78 ml). Using k-means clustering segmentation, the average tumor sizes calculated by the 2 observers were 2.31 ml (SD 1.83 ml) and 2.12 ml (SD 1.87 ml). The concordance correlation coefficient for the ellipsoid, GrowCut, and k-means clustering methods were 0.92 (95% CI, 0.83–0.99), 0.12 (95% CI, –0.21 to 0.44), and 0.95 (95% CI, 0.90–0.99), respectively. Conclusions: k-means clustering, a parameter-dependent method with low degree of subjective judgment, showed better interobserver agreement compared with the ellipsoid and GrowCut methods. k-means clustering clearly delineated tumor boundaries and outlined components of the tumor with different signal intensities.

Reliability of 3 Strategies of Orbital Tumor Volume Measurement Using Phantom Modeling

“…In a recent publication of volume measurement strategies for cavernous hemangiomas, EV had poor interobserver agreement and underestimated tumor volume as compared to the SA strategy. 8 We postulate that EV underestimated the size of ovoid tumors, due to the difficulty in selecting the longest diameter slices, and overestimated diffuse tumor size because selecting only 3 planes of measurements does not adequately reflect the irregular contour of the diffuse tumors. MS uses a slice-by-slice region of interest strategy to measure tumor volume.…”

“…[4][5][6][7] Prior studies that compare strategies to measure orbital tumor, fat, or EOM volume have failed to validate the results by measuring the true size of the tissue in vivo. 3,8 Even those few studies that have compared imaging derived tumor volume, in other parts of the body, with that measured in the resected pathologic specimen are flawed due to the significant volume changes that can occur after tumor removal from the host bed. 9 Moreover, most existing soft tissue measurement strategies employ CT, not higher resolution MRI.…”

Orbital Vascular Anomalies: A Nomenclatorial, Etiological, and Nosologic Conundrum

“…9 Of note is that it is not just older terminology that should be used with caution, but relatively newer terms popularized by ISSVA like "capillary malformation" (CM) and "venous malformation" (VM) should also be applied carefully to avoid further confusion. 14 In the case of VMs, the persistent and controversial use of the obsolete term "cavernous hemangioma" in the orbital 15 while being totally discarded by others 16 who paradoxically use the generic term VM instead of the more specific term (CVM) to describe these lesions. 16 This latter preference may be misleading as it makes no distinction between orbital CVMs and distensible VMs (the so-called orbital varices or common VMs in ISSVA 2018) and is probably based on a misinterpretation of the ISSVA classification which never listed orbital CVM in any of its iterations.…”

“…Although the term is generally falling out of favor by ISSVA proponents and is rapidly being replaced by the term cavernous VM (CVM), opinion is still divided in the oculoplastic literature. Most authors currently use the term orbital CVM, but the term cavernous hemangioma is still used by some authors, 15 while being totally discarded by others 16 who paradoxically use the generic term VM instead of the more specific term (CVM) to describe these lesions. 16 This latter preference may be misleading as it makes no distinction between orbital CVMs and distensible VMs (the so-called orbital varices or common VMs in ISSVA 2018) and is probably based on a misinterpretation of the ISSVA classification which never listed orbital CVM in any of its iterations.…”