Accurate Fine-Grained Segmentation of Human Anatomy in Radiographs via Volumetric Pseudo-Labeling

1 code implementation • 6 Jun 2023 • Constantin Seibold, Alexander Jaus, Matthias A. Fink, Moon Kim, Simon Reiß, Ken Herrmann, Jens Kleesiek, Rainer Stiefelhagen

Results: Our resulting segmentation models demonstrated remarkable performance on CXR, with a high average model-annotator agreement between two radiologists with mIoU scores of 0. 93 and 0. 85 for frontal and lateral anatomy, while inter-annotator agreement remained at 0. 95 and 0. 83 mIoU.

Anatomy Computed Tomography (CT) +2