Methods and systems are provided for identifying shape-based atypical segmentations. In one example, a method includes receiving a segmentation of a region of interest (ROI) of a medical image, the segmentation output by a segmentation model, calculating a confidence metric of the segmentation that indicates how well a shape of the segmentation can be encoded by an encoding of one or more principal modes of shape variation of a set of previously determined segmentations of the ROI, and responsive to the confidence metric meeting a predetermined condition relative to a threshold, displaying the segmentation, storing the segmentation, and/or using the segmentation in one or more downstream processes; otherwise, prompting a user to perform a manual segmentation.