loading page

KCB-Net: A 3D Knee Cartilage and Bone Segmentation Network via Sparse Annotation
  • +4
  • Yaopeng Peng ,
  • Hao Zheng ,
  • Fahim Zaman ,
  • Lichun Zhang ,
  • Xiaodong Wu ,
  • Milan Sonka ,
  • Danny Chen
Yaopeng Peng
University of Notre Dame

Corresponding Author:[email protected]

Author Profile
Hao Zheng
Author Profile
Fahim Zaman
Author Profile
Lichun Zhang
Author Profile
Xiaodong Wu
Author Profile
Milan Sonka
Author Profile
Danny Chen
Author Profile


Knee cartilage and bone segmentation is critical for physicians to analyze and diagnose articular damage and knee osteoarthritis (OA). Deep learning (DL) methods for medical image segmentation have largely outperformed traditional methods, but they often need large amounts of annotated data for model training, which is very costly and time-consuming for medical experts, especially on 3D images. In this paper, we report a new knee cartilage and bone segmentation framework, KCB-Net, for 3D MR images based on sparse annotation. KCB-Net selects a small subset of slices from 3D images for annotation, and seeks to bridge the performance gap between sparse annotation and full annotation. Specifically, it first identifies a subset of the most effective and representative slices with an unsupervised scheme; it then trains an ensemble model using the annotated slices; next, it self-trains the model using 3D images containing pseudo-labels generated by the ensemble method and improved by a bi-directional hierarchical earth mover’s distance (bi-HEMD) algorithm; finally, it fine-tunes the segmentation results using the primal-dual Internal Point Method (IPM). Experiments on two 3D MR knee joint datasets (the Iowa dataset and iMorphics dataset) show that our new framework outperforms state-of-the-art methods on full annotation, and yields high quality results even for annotation ratios as low as 5%.
Nov 2022Published in Medical Image Analysis volume 82 on pages 102574. 10.1016/j.media.2022.102574