S.K.: Incorporating priors on expert performance parameters for segmentation validation and label fusion: a maximum a posteriori staple

Abstract. In order to evaluate the quality of segmentations of an im-age and assess intra- and inter-expert variability in segmentation perfor-mance, an Expectation Maximization (EM) algorithm for Simultaneous Truth And Performance Level Estimation (STAPLE) was recently devel-oped. This algorithm, originally presented for segmentation validation, has since been used for many applications, such as atlas construction and decision fusion. However, the manual delineation of structures of in-terest is a very time consuming and burdensome task. Further, as the time required and burden of manual delineation increase, the accuracy of the delineation is decreased. Therefore, it may be desirable to ask the experts to delineate only a reduced number of structures or the seg-mentation of all structures by all experts may simply not be achieved. Fusion from data with some structures not segmented by each expert should be carried out in a manner that accounts for the missing informa-tion. In other applications, locally inconsistent segmentations may drive the STAPLE algorithm into an undesirable local optimum, leading to misclassifications or misleading experts performance parameters. We present a new algorithm that allows fusion with partial delineation and which can avoid convergence to undesirable local optima in the pres-ence of strongly inconsistent segmentations. The algorithm extends STA-PLE by incorporating prior probabilities for the expert performance pa-rameters. This is achieved through a Maximum A Posteriori formulation, where the prior probabilities for the performance parameters are mod-eled by a beta distribution. We demonstrate that this new algorithm enables dramatically improved fusion from data with partial delineation by each expert in comparison to fusion with STAPLE.