Robust Seed Mask Generation for Interactive Image Segmentation

TL;DR

This paper introduces an automatic seed generation method for interactive medical image segmentation, reducing initial user interaction time and improving segmentation accuracy with saliency-based seeding.

Contribution

The authors propose a novel automatic seeding pipeline using saliency recognition to bypass initial user input in medical image segmentation.

Findings

Median Dice score of 68.22% before user interaction

Error rate in seeding is only 0.088%

Reduces time-consuming initial interaction phase

Abstract

In interactive medical image segmentation, anatomical structures are extracted from reconstructed volumetric images. The first iterations of user interaction traditionally consist of drawing pictorial hints as an initial estimate of the object to extract. Only after this time consuming first phase, the efficient selective refinement of current segmentation results begins. Erroneously labeled seeds, especially near the border of the object, are challenging to detect and replace for a human and may substantially impact the overall segmentation quality. We propose an automatic seeding pipeline as well as a configuration based on saliency recognition, in order to skip the time-consuming initial interaction phase during segmentation. A median Dice score of 68.22% is reached before the first user interaction on the test data set with an error rate in seeding of only 0.088%.