Purpose: Comparison of imaging measurement devices in the absence of a gold-standard comparator remains a\nvexing problem; especially in scenarios where multiple, non-paired, replicated measurements occur, as in\nimage-guided radiotherapy (IGRT). As the number of commercially available IGRT presents a challenge to determine\nwhether different IGRT methods may be used interchangeably, an unmet need conceptually parsimonious and\nstatistically robust method to evaluate the agreement between two methods with replicated observations.\nConsequently, we sought to determine, using an previously reported head and neck positional verification dataset,\nthe feasibility and utility of a Comparison of Measurement Methods with the Mixed Effects Procedure Accounting for\nReplicated Evaluations (COM3PARE), a unified conceptual schema and analytic algorithm based upon Royââ?¬â?¢s linear\nmixed effects (LME) model with Kronecker product covariance structure in a doubly multivariate set-up, for IGRT\nmethod comparison.\nMethods: An anonymized dataset consisting of 100 paired coordinate (X/ measurements from a sequential series of\nhead and neck cancer patients imaged near-simultaneously with cone beam CT (CBCT) and kilovoltage X-ray (KVX)\nimaging was used for model implementation. Software-suggested CBCT and KVX shifts for the lateral (X), vertical (Y)\nand longitudinal (Z) dimensions were evaluated for bias, inter-method (between-subject variation), intra-method\n(within-subject variation), and overall agreement using with a script implementing COM3PARE with the MIXED\nprocedure of the statistical software package SAS (SAS Institute, Cary, NC, USA).\nResults: COM3PARE showed statistically significant bias agreement and difference in inter-method between CBCT\nand KVX was observed in the Z-axis (both pâË?â??value < 0.01). Intra-method and overall agreement differences were\nnoted as statistically significant for both the X- and Z-axes (all pâË?â??value < 0.01). Using pre-specified criteria, based on\nintra-method agreement, CBCT was deemed preferable for X-axis positional verification, with KVX preferred for\nsuperoinferior alignment.\nConclusions: The COM3PARE methodology was validated as feasible and useful in this pilot head and neck cancer\npositional verification dataset. COM3PARE represents a flexible and robust standardized analytic methodology for IGRT\ncomparison. The implemented SAS script is included to encourage other groups to implement COM3PARE in other\nanatomic sites or IGRT platforms.
Loading....