Back to all available

AI-driven multimodal diagnosis for knee osteoarthritis: A systematic review across task and fusion dimensions

algorithms-and-analysis-of-algorithms
artificial-intelligence
computer-vision
data-mining-and-machine-learning
neural-networks

Abstract

Background: Single-modality information could not fully reflect the pathological changes of knee osteoarthritis (KOA). This study systematically reviewed the research progress of AI-driven multimodal fusion methods for KOA diagnosis, using a “task type-fusion level” analytical framework.

Methodology: The study followed the PRISMA criteria and searched multiple databases for SCI-indexed publications from 2020 to 2025. Relevant and recent multimodal KOA studies were included. The selected literature was comparatively analyzed from multiple perspectives, including application classification, data modality, fusion method, and performance metrics.

Results: Kellgren-Lawrence (KL) grading was the most widely used application task in KOA research (over 50%). However, research directions gradually expanded to include cartilage damage detection and total knee replacement (TKR) prediction. Feature-level fusion was the most frequently applied strategy in these studies, demonstrating greater stability. Furthermore, attention mechanisms and transformer-based architectures showed promising potential in fusion modeling.

Conclusion: This study summarized the differences in AI-driven multimodal KOA research regarding fusion strategies and application directions. Although existing studies significantly improved diagnostic accuracy and predictive capability, challenges remained, including data standardization, cross-center generalization, and privacy protection. Future research should integrate time-series modeling and federated learning to develop more interpretable and clinically applicable KOA diagnostic models.

Ask to review this manuscript

Notes for potential reviewers

  • Volunteering is not a guarantee that you will be asked to review. There are many reasons: reviewers must be qualified, there should be no conflicts of interest, a minimum of two reviewers have already accepted an invitation, etc.
  • This is NOT OPEN peer review. The review is single-blind, and all recommendations are sent privately to the Academic Editor handling the manuscript. All reviews are published and reviewers can choose to sign their reviews.
  • What happens after volunteering? It may be a few days before you receive an invitation to review with further instructions. You will need to accept the invitation to then become an official referee for the manuscript. If you do not receive an invitation it is for one of many possible reasons as noted above.
  • PeerJ Computer Science does not judge submissions based on subjective measures such as novelty, impact or degree of advance. Effectively, reviewers are asked to comment on whether or not the submission is scientifically and technically sound and therefore deserves to join the scientific literature. Our Peer Review criteria can be found on the "Editorial Criteria" page - reviewers are specifically asked to comment on 3 broad areas: "Basic Reporting", "Experimental Design" and "Validity of the Findings".
  • Reviewers are expected to comment in a timely, professional, and constructive manner.
  • Until the article is published, reviewers must regard all information relating to the submission as strictly confidential.
  • When submitting a review, reviewers are given the option to "sign" their review (i.e. to associate their name with their comments). Otherwise, all review comments remain anonymous.
  • All reviews of published articles are published. This includes manuscript files, peer review comments, author rebuttals and revised materials.
  • Each time a decision is made by the Academic Editor, each reviewer will receive a copy of the Decision Letter (which will include the comments of all reviewers).

If you have any questions about submitting your review, please email us at [email protected].