Journal article
KCB-Net: A 3D knee cartilage and bone segmentation network via sparse annotation
Medical image analysis, Vol.82, pp.102574-102574
09/07/2022
DOI: 10.1016/j.media.2022.102574
PMCID: PMC10515734
PMID: 36126403
Abstract
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 four 3D MR knee joint datasets (the SKI10 dataset, OAI ZIB dataset, Iowa dataset, and iMorphics dataset) show that our new framework outperforms state-of-the-art methods on full annotation, and yields high quality results for small annotation ratios even as low as 10%.
Details
- Title: Subtitle
- KCB-Net: A 3D knee cartilage and bone segmentation network via sparse annotation
- Creators
- Yaopeng PengHao ZhengPeixian LiangLichun ZhangFahim ZamanXiaodong WuMilan SonkaDanny Z Chen
- Resource Type
- Journal article
- Publication Details
- Medical image analysis, Vol.82, pp.102574-102574
- DOI
- 10.1016/j.media.2022.102574
- PMID
- 36126403
- PMCID
- PMC10515734
- NLM abbreviation
- Med Image Anal
- eISSN
- 1361-8423
- Grant note
- DOI: 10.13039/100000001, name: National Science Foundation, award: CCF-1617735; DOI: 10.13039/100000070, name: National Institute of Biomedical Imaging and Bioengineering, award: R01-EB004640
- Language
- English
- Date published
- 09/07/2022
- Academic Unit
- Roy J. Carver Department of Biomedical Engineering; Electrical and Computer Engineering; Iowa Technology Institute; Radiation Oncology; The Iowa Institute for Biomedical Imaging; Fraternal Order of Eagles Diabetes Research Center; Injury Prevention Research Center; Ophthalmology and Visual Sciences
- Record Identifier
- 9984297158602771
Metrics
20 Record Views