Deep learning-enabled segmentation of knee cartilage in conventional magnetic resonance images: Internal and external validation of different models
The Knee · 2026-01 · 58 · 104270
DOI: 10.1016/j.knee.2025.10.024 ↗Kurz gefasst
Validierungsstudie in der Zeitschrift The Knee, 2026-01. Verfasst von Lin Yao und 10 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiology, the Fifth Affiliated Hospital of Sun Yat-sen University, Zhuhai, Guangdong Province, China. Hauptthemen laut MEDLINE: Deep Learning, Cartilage, Articular, Magnetic Resonance Imaging, Knee Joint. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 36 Literaturangaben. Gefördert von: Medical Science and Technology Foundation of Guangdong Province, National Natural Science Foundation of China.
Abstract (Originalsprache)
BACKGROUND: Accurate evaluation of the cartilage anatomy of the knee is helpful for clinical evaluation of the source of knee pain and the classification and treatment of knee osteoarthritis (OA). This study proposes a deep learning model for segmentation of knee articular cartilage in conventional proton density fat-saturated MRI sequences to assess cartilage morphology for subsequent injury grading. METHODS: This retrospective study was conducted at two radiology centers, involving 254 knees from 254 patients who had previously undergone MRI scans. The training-internal validation cohort included 219 knees from Center 1. The external validation cohort comprised 35 knees from Center 2. Two musculoskeletal radiology experts manually annotated the cartilage regions. A 3D Res U-net model was employed for segmentation, and its performance was compared with 3D U-net and 3D V-net models. Segmentation results were evaluated using the Dice coefficient and Jaccard index. RESULTS: The 3D Res U-net model demonstrated superior segmentation performance compared to the other deep learning methods. For cartilage in the lateral femorotibial joint, medial femorotibial joint, and patellofemoral joint, the average Dice coefficients with 3D Res U-net were 0.871, 0.860, and 0.858 in internal validation and 0.846, 0.837, and 0.819 in external validation, respectively. The Jaccard index followed a similar trend. CONCLUSION: The 3D Res U-net model improves knee cartilage segmentation in conventional MR imaging, contributing to the understanding of cartilage morphology and the improvement of clinically relevant decisions.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Radiology, the Fifth Affiliated Hospital of Sun Yat-sen University, Zhuhai, Guangdong Province, China.
- Department of Spinal Surgery, the Fifth Affiliated Hospital of Sun Yat-Sen University, Zhuhai, Guangdong Province, China.
- Department of Radiology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China.
- Huiying Medical Technology (Beijing), Huiying Medical Technology Co. Ltd, Beijing, China.
- Department of Radiology, the Fourth People's Hospital of Guiyang, Guiyang, Guizhou Province, China.
- Department of Medical Imaging, Foshan Hospital of Traditional Chinese Medicine, Foshan City, Guangdong Province, China.
Förderung
- Medical Science and Technology Foundation of Guangdong Province
- National Natural Science Foundation of China
Zitieren
APA
Yao, L., Liang, X., Liu, J., Qian, B., Xiao, Z., Zhu, D., Feng, J., Zhou, S., Cui, H., Li, S., & Li, W. (2026). Deep learning-enabled segmentation of knee cartilage in conventional magnetic resonance images: Internal and external validation of different models. The Knee, 58, 104270. https://doi.org/10.1016/j.knee.2025.10.024
Vancouver
Yao L, Liang X, Liu J, Qian B, Xiao Z, Zhu D, et al. Deep learning-enabled segmentation of knee cartilage in conventional magnetic resonance images: Internal and external validation of different models. The Knee. 2026;58:104270. doi:10.1016/j.knee.2025.10.024
BibTeX
@article{Yao2026,
author = {Yao, Lin and Liang, Xiaoling and Liu, Jin and Qian, Baoxin and Xiao, Zhongli and Zhu, Dantian and Feng, Jiaxin and Zhou, Shouguo and Cui, Huaqian and Li, Shaolin and Li, Wei},
title = {Deep learning-enabled segmentation of knee cartilage in conventional magnetic resonance images: Internal and external validation of different models},
journal = {The Knee},
year = {2026},
volume = {58},
pages = {104270},
doi = {10.1016/j.knee.2025.10.024}
}
EUVIMED