OsteoFusionFormer: dual-stage transformer fusion framework for knee osteoporosis diagnosis
The Knee · 2026-08 · 61 · 104428
DOI: 10.1016/j.knee.2026.104428 ↗Kurz gefasst
Fachartikel in der Zeitschrift The Knee, 2026-08. Verfasst von Eshika Jain und 2 weiteren Autor:innen. Beteiligte Einrichtung: Centre for Research Impact and Outcome, Chitkara University Institute of Engineering and Technology, Chitkara University, Punjab, India. Electronic a…. Hauptthemen laut MEDLINE: Osteoporosis, Knee Joint. Die Arbeit wurde bisher 22-mal zitiert (Crossref) und nennt 33 Literaturangaben.
Abstract (Originalsprache)
AIM: The aim of the study is to introduce a new model, OsteoFusionFormer, namely a dual transformer model for automatic classification of knee osteoporosis into three groups: Normal, Osteopenia, and Osteoporosis. The objective was to overcome single-branch transformer limitations by incorporating anatomical global context and fine-grained bone features to increase diagnostic accuracy. METHOD: OsteoFusionFormer combines two parallel arms, a Vision Transformer (ViT) for global anatomical representation and a Bone-Aware Transformer (BAT) for localised bone-specific features. These are combined with a hierarchical dual fusion strategy. First, cross-attention enables feature-level fusion between ViT and BAT embeddings. Next, a confidence-weighted decision-level fusion employs an auxiliary gating network to compute adaptive weights (α1, α2), yielding a soft-ensemble prediction. Ablation studies systematically remove modules (ViT, BAT, gating, CLAHE) to assess contributions. Interpretability is assessed via attention maps and Grad-CAM++. RESULTS: OsteoFusionFormer had 96.8% overall accuracy, which exceeds the accuracy of ViT-only (91.3%), BAT-only (90.1%), and late fusion averaging (93.6%). Ablation verified a drop in performance without BAT (-5.5%), ViT (-6.7%), gating (-3.2%), and CLAHE (-4.4%). Performance was verified with 15 new kinds of bone-specific indicators: Bone-Aware Accuracy: 96.8%, Trabecular Sensitivity Index: 95.2%, Cortical Degeneration Detection Rate: 97.6%, Joint Space Narrowing Recall: 96.1%, Bone Class Specificity: 97.2%, Osteopenia Detection Precision: 92.4%, Bone Focus Ratio: 91.8%, Bone Entropy Index: 0.26 bits, Visual interpretability showed good expert agreement (BIAS: 87.3%). CONCLUSION: By combining global and local bone features, OsteoFusionFormer provides better accuracy, diagnosis sensitivity, and structure focus with an explainability guarantee.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Centre for Research Impact and Outcome, Chitkara University Institute of Engineering and Technology, Chitkara University, Punjab, India. Electronic address: eshikajain0150@gmail.com.
- Centre for Research Impact and Outcome, Chitkara University Institute of Engineering and Technology, Chitkara University, Punjab, India. Electronic address: onlyvinaykukreja@gmail.com.
- Centre for Research Impact and Outcome, Chitkara University Institute of Engineering and Technology, Chitkara University, Punjab, India. Electronic address: prathamkaushik8@gmail.com.
Zitieren
APA
Jain, E., Kukreja, V., & Kaushik, P. (2026). OsteoFusionFormer: dual-stage transformer fusion framework for knee osteoporosis diagnosis. The Knee, 61, 104428. https://doi.org/10.1016/j.knee.2026.104428
Vancouver
Jain E, Kukreja V, Kaushik P. OsteoFusionFormer: dual-stage transformer fusion framework for knee osteoporosis diagnosis. The Knee. 2026;61:104428. doi:10.1016/j.knee.2026.104428
BibTeX
@article{Jain2026,
author = {Jain, Eshika and Kukreja, Vinay and Kaushik, Pratham},
title = {OsteoFusionFormer: dual-stage transformer fusion framework for knee osteoporosis diagnosis},
journal = {The Knee},
year = {2026},
volume = {61},
pages = {104428},
doi = {10.1016/j.knee.2026.104428}
}
EUVIMED