scholarly journals Histogram Based Synovitis Scoring System Using Ultrasound Images of Rheumatoid Arthritis

2018 ◽  
Author(s):  
RJ Hemalatha ◽  
V Vijayabaskarin
Keyword(s):  
Rheumatoid Arthritis ◽  
Scoring System ◽  
Ultrasound Images

scholarly journals OP0064 AUTOMATIC SCORING OF ARTHRITIS DISEASE ACTIVITY ON ULTRASOUND IMAGES FROM RHEUMATOID ARTHRITIS PATIENTS WITH CASCADED CONVOLUTIONAL NEURAL NETWORKS

2020 ◽  
Vol 79 (Suppl 1) ◽  
pp. 42.2-43
Author(s):  
A. Christensen ◽  
S. A. Just ◽  
J. K. H. Andersen ◽  
T. R. Savarimuthu
Keyword(s):  
Neural Network ◽  
Rheumatoid Arthritis ◽  
Neural Networks ◽  
Early Detection ◽  
Disease Activity ◽  
Convolutional Neural Networks ◽  
Scoring System ◽  
Cascade Model ◽  
Ultrasound Images ◽  
Automatic Scoring

Background:Systematic Power or Color Doppler (CD) ultrasound (US) of joints can be used for early detection of Rheumatoid Arthritis (RA), predicting radiographic progression and early detection of disease flare in established RA [1, 2]. The international standard for performing RA US scanning and evaluation of disease activity is the OMERACT-ELUAR Synovitis Scoring (OESS) system [1, 3].To further mitigate the operator-dependency in scoring disease activity on CD US images in future trials and clinical practice, we proposed the use of convolutional neural networks (CNN) to automatically grade CD US images according to the OESS definitions. This study is a continuation of the findings in our previous work, where we developed a CNN for four-class CD US OESS scoring with a test accuracy of 75.0% [4].Objectives:Since our last contribution, we have further developed the architecture of this neural network and can here present a new idea applying a Cascaded Convolutional Neural Network design. We evaluate the generalizability of this method on unseen data, comparing the CNN with an expert rheumatologist.Methods:The images used for developing the algorithms were graded by a single expert rheumatologist according to the OESS system. The CNNs in the cascade were trained individually, after which they were combined to form the cascade model as shown in figure 1. The algorithms were evaluated on a separate test dataset, which came from the same distribution as the training dataset. The algorithms were compared to the gradings of an expert rheumatologist on a per-joint basis using a Kappa test, and on a per-patient basis using a Wilcoxon Signed Rank test.Figure 1.CNN-1 is the first CNN in the model and distinguishes between RA disease grade (DG) 0 and DG’s 1, 2 and 3. CNN-2 is the second CNN and distinguishes between DG 1 and DG’s 2 and 3. CNN-3 is the final CNN which distinguishes between DG’s 2 and 3.Results:With 1678 images available for training and 322 images for testing the model, the model achieved an overall 4-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).Conclusion:We have shown that dividing a four-degree classification task into three successive binary classification tasks has resulted in a model capable of making correct classifications in 83.9% of the cases for a test set of ultrasound images with a naturally occurring distribution of RA joint disease activity scores.Furthermore, we have shown that the cascade model can produce classification decisions comparable with a human rheumatologist when applied on a per-patient basis. This emphasizes the potential of using CNNs with this architecture as a strong assistive tool for the objective assessment of disease activity of RA patients.References:[1]D’Agostino M-A, Terslev L, Aegerter P, et al. (2017). Scoring ultrasound synovitis in rheumatoid arthritis: a OMERACT-EULAR ultrasound taskforce-Part 1: definition and development of a standardised, consensus-based scoring system. RMD Open.[2]Paulshus NS, Aga A-B, Olsen I, et al. (2018). Clinical and ultrasound remission after 6 months of treat-to-target therapy in early rheumatoid arthritis: associations to future good radiographic and physical outcomes. Ann Rheum Dis, 77, s. 1425-25.[3]Terslev L, Naredo E, Aegerter P, et al. (3 2017). Scoring ultrasound synovitis in rheumatoid arthritis: a OMERACT-EULAR ultrasound taskforce-Part 2: reliability and application to multiple joints of a standardised consensus-based scoring system. RMD Open.[4]Andersen JKH, Pedersen JS, Laursen MS, et al. Neural networks for automatic scoring of arthritis disease activity on ultrasound images. RMD Open 2019; 5:e000891. doi:10.1136/ rmdopen-2018-000891Disclosure of Interests:None declared

The Reliability of Ultrasound in the Assessment of Hyaline Cartilage in Rheumatoid Arthritis and Healthy Metacarpal Heads

2020 ◽  
Author(s):  
Edoardo Cipolletta ◽  
Emilio Filippucci ◽  
Andrea Di Matteo ◽  
Giulia Tesei ◽  
Micaela Ana Cosatti ◽  
...  
Keyword(s):  
Rheumatoid Arthritis ◽  
Scoring System ◽  
Quantitative Assessment ◽  
Hyaline Cartilage ◽  
Scoring Systems ◽  
Thickness Measurement ◽  
Cartilage Thickness ◽  
Observer Agreement ◽  
Very High Frequency ◽  
Observer Reliability

Abstract Purpose i) To assess the inter- and intra-observer reliability of ultrasound (US) in the evaluation of the hyaline cartilage (HC) of the metacarpal head (MH) in patients with rheumatoid arthritis (RA) and in healthy subjects (HS) both qualitatively and quantitatively. ii) To calculate the smallest detectable difference (SDD) of the MH cartilage thickness measurement. iii) To correlate the qualitative scoring system and the quantitative assessment. Materials and Methods US examination was performed on 280 MHs of 20 patients with RA and 15 HS using a very high frequency probe (up to 22 MHz). HC status was evaluated both qualitatively (using a five-grade scoring system) and quantitatively (using the average value of the longitudinal and transverse measures). The HC of MHs from II to V metacarpophalangeal joint of both hands were scanned independently on the same day by two rheumatologists to assess inter-observer reliability. All subjects were re-examined using the same scanning protocol and the same US setting by one sonographer after a week to assess intra-observer reliability. Results The inter-observer agreement and intra-observer agreement were moderate to substantial (k = 0.66 and k = 0.73) for the qualitative scoring system and high (ICC = 0.93 and ICC = 0.94) for the quantitative assessment. The SDD of the MH cartilage thickness measurement was 0.09 mm. A significant correlation between the two scoring systems was found (r = –0.35; p < 0.001). Conclusion The present study describes the main methodological issues of HC assessment. Using a standardized protocol, both the qualitative and the quantitative scoring systems can be reliable.

Value of Ultrasound Images and Reports Scoring System in Quality Control

2021 ◽  
Vol 5 (3) ◽  
pp. 245
Author(s):  
Qiu, MD Li ◽  
Peng, MD Yulan ◽  
Lu, MD Qiang ◽  
Luo, MD Yan
Keyword(s):  
Quality Control ◽  
Scoring System ◽  
Ultrasound Images

A Novel Fluorescence Optical Imaging Scoring System for Hand Synovitis in Rheumatoid Arthritis – validity and agreement with ultrasound

Rheumatology ◽  
2021 ◽  
Author(s):  
Mads Ammitzbøll Danielsen ◽  
Daniel Glinatsi ◽  
Lene Terslev ◽  
Mikkel Østergaard
Keyword(s):  
Rheumatoid Arthritis ◽  
Optical Imaging ◽  
Scoring System ◽  
Moderate Correlation ◽  
Treatment Intensification ◽  
Fluorescence Optical Imaging ◽  
Reliable Assessment ◽  
Intraclass Correlations ◽  
Reader Reliability ◽  
Foie Gras

Abstract Objectives To develop and validate a new semiquantitative Fluorescence Optical Imaging (FOI) scoring system – the FOI Enhancement-Generated rheumatoid arthritis (RA) Score (FOIE-GRAS) for synovitis assessment in the hand. Methods The development of FOIE-GRAS was based on consensus of 4 experts in musculoskeletal imaging. Forty-six RA patients, eligible for treatment intensification and with ≥1 clinically swollen joint in the hands and 11 healthy controls were included. FOI, ultrasound and clinical assessment of both hands were obtained at baseline and for RA patients after 3- and 6-months’ follow-up. Twenty RA patients had an FOI rescan after 4 hours. Synovitis was scored using FOIE-GRAS and the OMERACT ultrasound synovitis scoring system. All FOI images were scored by 2 readers. Inter-scan, inter-and intra-reader reliability were determined. Furthermore, FOIE-GRAS agreement with ultrasound and responsiveness was assessed. Results FOIE-GRAS synovitis was defined as early enhancement and scores based on the degree of coverage of the specific joint region after 3 seconds (0–3). Inter-scan, intra- and inter-reader intraclass correlations coefficients (ICC) were good-excellent for all baseline scores (0.76-0.98) and moderate-to-good for change (0.65-76). The FOIE-GRAS had moderate agreement with ultrasound (ICC 0.30-0.54) for total score, a good standardized response mean (&gt;0.80), and moderate correlation with clinical joint assessment and DAS28-CRP. The median (IQR) reading time per FOI examination was 133 (109;161) seconds. Scores were significantly lower in controls 1(0;4) than RA patients 11(6;19). Conclusion The FOIE-GRAS offers a feasible and reliable assessment of synovitis in RA, with a moderate correlation with ultrasound and DAS28CRP, and good responsiveness.

scholarly journals The OMERACT Rheumatoid Arthritis Magnetic Resonance Imaging (MRI) Scoring System: Updated Recommendations by the OMERACT MRI in Arthritis Working Group

2017 ◽  
Vol 44 (11) ◽  
pp. 1706-1712 ◽  
Author(s):  
Mikkel Østergaard ◽  
Charles G. Peterfy ◽  
Paul Bird ◽  
Frédérique Gandjbakhch ◽  
Daniel Glinatsi ◽  
...  
Keyword(s):  
Rheumatoid Arthritis ◽  
Magnetic Resonance Imaging ◽  
Magnetic Resonance ◽  
Working Group ◽  
Scoring System ◽  
Research Agenda ◽  
Image Resolution ◽  
Sensitivity To Change ◽  
Resonance Imaging ◽  
Magnetic Resonance Imaging Mri

Objective.The Outcome Measures in Rheumatology (OMERACT) Rheumatoid Arthritis (RA) Magnetic Resonance Imaging (MRI) scoring system (RAMRIS), evaluating bone erosion, bone marrow edema/osteitis, and synovitis, was introduced in 2002, and is now the standard method of objectively quantifying inflammation and damage by MRI in RA trials. The objective of this paper was to identify subsequent advances and based on them, to provide updated recommendations for the RAMRIS.Methods.MRI studies relevant for RAMRIS and technical and scientific advances were analyzed by the OMERACT MRI in Arthritis Working Group, which used these data to provide updated considerations on image acquisition, RAMRIS definitions, and scoring systems for the original and new RA pathologies. Further, a research agenda was outlined.Results.Since 2002, longitudinal studies and clinical trials have documented RAMRIS variables to have face, construct, and criterion validity; high reliability and sensitivity to change; and the ability to discriminate between therapies. This has enabled RAMRIS to demonstrate inhibition of structural damage progression with fewer patients and shorter followup times than has been possible with conventional radiography. Technical improvements, including higher field strengths and improved pulse sequences, allow higher image resolution and contrast-to-noise ratio. These have facilitated development and validation of scoring methods of new pathologies: joint space narrowing and tenosynovitis. These have high reproducibility and moderate sensitivity to change, and can be added to RAMRIS. Combined scores of inflammation or joint damage may increase sensitivity to change and discriminative power. However, this requires further research.Conclusion.Updated 2016 RAMRIS recommendations and a research agenda were developed.

Diagnostic Imaging Support System for Rheumatoid Arthritis Using Ultrasound Images

2021 ◽  
Author(s):  
Kenichi Arai ◽  
Chisato Miura ◽  
Shin-Ya Kawashiri ◽  
Tetsuo Imai ◽  
Toru Kobayashi
Keyword(s):  
Rheumatoid Arthritis ◽  
Diagnostic Imaging ◽  
Support System ◽  
Ultrasound Images

Applying cascaded convolutional neural network design further enhances automatic scoring of arthritis disease activity on ultrasound images from rheumatoid arthritis patients

2020 ◽  
Vol 79 (9) ◽  
pp. 1189-1193
Author(s):  
Anders Bossel Holst Christensen ◽  
Søren Andreas Just ◽  
Jakob Kristian Holm Andersen ◽  
Thiusius Rajeeth Savarimuthu
Keyword(s):  
Neural Network ◽  
Rheumatoid Arthritis ◽  
Disease Activity ◽  
Convolutional Neural Network ◽  
Objective Assessment ◽  
Rank Test ◽  
Ultrasound Images ◽  
Data Set ◽  
Unseen Data ◽  
Significant Difference

ObjectivesWe 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.MethodsThe 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.ResultsWith 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%.ConclusionsUsing 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.

Sign in / Sign up

Export Citation Format

Share Document