Validation of Medical Image Processing in Image-Guided Therapy

CLINICAL use of image-guided therapy (IGT) systems is widespread today and growing rapidly, creating the need for a common and rigorous validation methodology, as reported in recent workshops and conferences [1]–[6]. One key characteristic of IGT systems is that they employ medical image processing methods (e.g., segmentation, registration, visualization, and calibration). As a result of this intrinsic structure, validation of IGT systems should include both individual components, the overall system and evaluation of how uncertainties propagate through the entire IGT process. Today, almost all peerreviewed publications reporting on the development of new medical image processing methods include a validation section, although this was not always true in the past. Validation of a medical image processing method allows its intrinsic characteristics to be highlighted, as well as evaluation of its performance and limitations. Moreover, validation clarifies the potential clinical contexts or applications that the method may serve. Validation may also demonstrate a method’s clinical added value as well as estimate social or economic impact. However, standardization of validation processes is required in order to compare various IGT systems. Validation tests can facilitate the user’s task of determining whether a particular system meets a given set of clinical requirements. This editorial identifies the principal requirements of IGT system validation and encourages the medical imaging community to develop a common methodology and terminology so we may all share analyses and results in this topic.