Article Text

Download PDFPDF
Applying cascaded convolutional neural network design further enhances automatic scoring of arthritis disease activity on ultrasound images from rheumatoid arthritis patients
  1. Anders Bossel Holst Christensen1,
  2. Søren Andreas Just2,
  3. Jakob Kristian Holm Andersen1,
  4. Thiusius Rajeeth Savarimuthu1
  1. 1 Maersk Mc-Kinney Moller Institute, University of Southern Denmark, Odense, Denmark
  2. 2 Department of Rheumatology, Odense University Hospital, Odense, Denmark
  1. Correspondence to Anders Bossel Holst Christensen, Maersk Mc-Kinney Moller Institute, University of Southern Denmark, Odense 5230, Denmark; abc{at}


Objectives We have previously shown that neural network technology can be used for scoring arthritis disease activity in ultrasound images from rheumatoid arthritis (RA) patients, giving scores according to the EULAR-OMERACT grading system. We have now further developed the architecture of this neural network and can here present a new idea applying cascaded convolutional neural network (CNN) design with even better results. We evaluate the generalisability of this method on unseen data, comparing the CNN with an expert rheumatologist.

Methods The images were graded by an expert rheumatologist according to the EULAR-OMERACT synovitis scoring system. CNNs were systematically trained to find the best configuration. The algorithms were evaluated on a separate test data set and compared with the gradings of an expert rheumatologist on a per-joint basis using a Kappa statistic, and on a per-patient basis using a Wilcoxon signed-rank test.

Results With 1678 images available for training and 322 images for testing the model, it achieved an overall four-class accuracy of 83.9%. On a per-patient level, there was no significant difference between the classifications of the model and of a human expert (p=0.85). Our original CNN had a four-class accuracy of 75.0%.

Conclusions Using a new network architecture we have further enhanced the algorithm and have shown strong agreement with an expert rheumatologist on a per-joint basis and on a per-patient basis. This emphasises the potential of using CNNs with this architecture as a strong assistive tool for the objective assessment of disease activity of RA patients.

  • rheumatoid arthritis
  • disease activity
  • ultrasonography

Statistics from


  • Handling editor Josef S Smolen

  • Twitter @JustSoren

  • Contributors Development, training and evaluation of algorithms: ABHC. Collection and grading of data: SAJ. Design of the study: ABHC, TRS, SAJ. Counselling: TRS, SAJ, JKHA. Drafting, revising and final approval: ABHC, SAJ, TRS.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Disclaimer This study was done as a master thesis project at the Maersk Mc-Kinney Moller Institute, University of Southern Denmark, and has not received any funding.

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting or dissemination plans of this research.

  • Patient consent for publication Not required.

  • Ethics approval The SynRA study, where ultrasound images are from, is approved by the regional ethics review board (S-20140062) and the Danish data protection agency (2008-58-0035). All participants gave oral and written consent to participate.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data availability statement No data are available.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.