Semi-Supervised Knee Cartilage Segmentation With Successive Eigen Noise-Assisted Mean Teacher Knowledge Distillation

• Published in: IEEE transactions on medical imaging Vol. 44; no. 7; pp. 3051 - 3063
• Main Authors: Khan, Sheheryar, Ammar Khawer, Muhammad, Qureshi, Rizwan, Nawaz, Mehmood, Asim, Muhammad, Chen, Weitian, Yan, Hong
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.07.2025
• Subjects: Adaptation models; Algorithms; Annotations; Cartilage, Articular - diagnostic imaging; cross-modality; Data models; deep learning; domain adaptation; Feature extraction; Humans; Image Interpretation, Computer-Assisted - methods; Image segmentation; Knee Joint - diagnostic imaging; knowledge distillation; Magnetic resonance imaging; Magnetic Resonance Imaging - methods; medical image segmentation; multi-domain model; Noise; Osteoarthritis, Knee - diagnostic imaging; Subspace; Supervised Machine Learning; Three-dimensional displays; Training; Translation
• ISSN: 0278-0062; 1558-254X; 1558-254X
• DOI: 10.1109/TMI.2025.3556870

Knee cartilage segmentation for Knee Osteoarthritis (OA) diagnosis is challenging due to domain shifts from varying MRI scanning technologies. Existing cross-modality approaches often use paired order matching or style translation techniques to align features. Still, these methods can sacrifice discrimination in less prominent cartilages and overlook critical higher-order correlations and semantic information. To address this issue, we propose a novel framework called Successive Eigen Noise-assisted Mean Teacher Knowledge Distillation (SEN-MTKD) for adapting 2D knee MRI images across different modalities using partially labeled data. Our approach includes the Eigen Low-rank Subspace (ELRS) module, which employs low-rank approximations to generate meaningful pseudo-labels from domain-invariant feature representations progressively. Complementing this, the Successive Eigen Noise (SEN) module introduces advanced data perturbation to enhance discrimination and diversity in small cartilage classes. Additionally, we propose a subspace-based feature distillation loss mechanism (LRBD) to manage variance and leverage rich intermediate representations within the teacher model, ensuring robust feature representation and labeling. Our framework identifies a mutual cross-domain subspace using higher-order structures and lower energy latent features, providing reliable supervision for the student model. Extensive experiments on public and private datasets demonstrate the effectiveness of our method over state-of-the-art benchmarks. The code is available at github.com/AmmarKhawer/SEN-MTKD.