FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4379649137> ?p ?o ?g. }

• W4379649137 endingPage "1286" @default.
• W4379649137 startingPage "1286.1" @default.
• W4379649137 abstract "Background Determining the response to early treatment of Rheumatoid Arthritis (RA) or the at-risk stages for RA from MRI scans requires sensitive and specific measurements of imaging biomarkers of inflammation. Employing artificial intelligence (AI), especially deep learning, can be an option to measure RA treatment effects. Objectives We aimed to assess the ability of deep learning to determine treatment response by distinguishing MRIs of the wrist between treatment and placebo arms (as proxy for treatment effects) from the TREAT-EARLIER trial. Methods Wrist MRIs (contrast enhanced T1-weighted TSE fat suppressed sequences) were collected from 236 patients with clinically suspect arthralgia at four time points (baseline, with 4, 12 and 24 months follow-up) to determine the response to treatment by intramuscular methylprednisolone followed by methotrexate during one year [1]. 3D wrist MRI data were reconstructed in super-resolution from axial and coronal images. Since a statistically significant treatment effect was determined previously y RAMRIS [1], we used these 3D wrist MRI data from baseline and after 4 months to classify patients into treatment arms by a 3D convolutional neural network. Five different inputs were explored for training the model: 1) a difference image (baseline image simply subtracted from follow-up); 2) the combination of baseline and follow-up image; 3) the combination of baseline and difference image; 4) change maps (AI-based maps containing changes between two time points); and 5) the combination of baseline and change maps (see bottom panel in Figure 1). To evaluate the proposed model, 10-fold cross-validation was repeated five times with five different splits, and the area under the receiver operator curve (AUC) was reported. Results The mean ( ± SD) AUCs are presented in Table 1. As shown, the combination of different MRIs obtained promising results. Specifically the combination of baseline images and AI-based extracted change maps could improve the accuracy. Compared to a setting where follow-up and baseline scans were simply ‘subtracted’, this improved the prediction of treatment from 0.72 to 0.76. Table 1. Obtained AUC using Wrist MRI. (BL: baseline MRI; FU: Follow-up MRI; Diff: Difference image; and CM: Change Map) Input AUC (mean ± SD) Diff (= BL-FU) 0.71 ± 0.027 BL and FU 0.70 ± 0.032 BL and Diff 0.72 ± 0.017 CM 0.73 ± 0.009 BL and CM 0.76 ± 0.008 Conclusion The results show that deep learning models using baseline and follow-up MRI specifically AI-extracted change maps can accurately determine treatment response in CSA patients. Reference [1]Krijbolder, DI, et al. “Intervention with methotrexate in patients with arthralgia at risk of rheumatoid arthritis to reduce the development of persistent arthritis and its disease burden (TREAT EARLIER): a randomised, double-blind, placebo-controlled, proof-of-concept trial.” The Lancet 400.10348 (2022): 283-294. Figure 1. Overview of the proposed model. Acknowledgements This AI-project has been funded by the Dutch Research Council (NWO) Applied and Engineering Sciences (project number 17970). The TREAT-EALRIER trial has been funded by an NWO-ZonMW grant (project number 95104004). The Dutch Arthritis Society contributed financially to both grants. Disclosure of Interests Tahereh Hassanzadeh Grant/research support from: Bristol-Myers Squibb and Pfizer contributed to this project, through a grant from the Dutch Research Council (NWO), Applied and Engineering Sciences., Denis Shamonin Grant/research support from: Bristol-Myers Squibb and Pfizer contributed to this project, through a grant from the Dutch Research Council (NWO), Applied and Engineering Sciences., Yanli Li: None declared, Monique Reijnierse Grant/research support from: Bristol-Myers Squibb and Pfizer contributed to this project, through a grant from the Dutch Research Council (NWO), Applied and Engineering Sciences., Annette van der Helm – van Mil Grant/research support from: Bristol-Myers Squibb and Pfizer contributed to this project, through a grant from the Dutch Research Council (NWO), Applied and Engineering Sciences., Berend Stoel Grant/research support from: Bristol-Myers Squibb and Pfizer contributed to this project, through a grant from the Dutch Research Council (NWO), Applied and Engineering Sciences." @default.
• W4379649137 created "2023-06-08" @default.
• W4379649137 creator A5003620746 @default.
• W4379649137 creator A5027773478 @default.
• W4379649137 creator A5044797355 @default.
• W4379649137 creator A5056059876 @default.
• W4379649137 creator A5071884339 @default.
• W4379649137 creator A5087940439 @default.
• W4379649137 date "2023-05-30" @default.
• W4379649137 modified "2023-09-26" @default.
• W4379649137 title "AB0205 RA TREATMENT EFFECTS IN WRIST MRIS, DETERMINED BY DEEP LEARNING" @default.
• W4379649137 doi "https://doi.org/10.1136/annrheumdis-2023-eular.3600" @default.
• W4379649137 hasPublicationYear "2023" @default.
• W4379649137 type Work @default.
• W4379649137 citedByCount "0" @default.
• W4379649137 crossrefType "journal-article" @default.
• W4379649137 hasAuthorship W4379649137A5003620746 @default.
• W4379649137 hasAuthorship W4379649137A5027773478 @default.
• W4379649137 hasAuthorship W4379649137A5044797355 @default.
• W4379649137 hasAuthorship W4379649137A5056059876 @default.
• W4379649137 hasAuthorship W4379649137A5071884339 @default.
• W4379649137 hasAuthorship W4379649137A5087940439 @default.
• W4379649137 hasBestOaLocation W43796491371 @default.
• W4379649137 hasConcept C126322002 @default.
• W4379649137 hasConcept C126838900 @default.
• W4379649137 hasConcept C142724271 @default.
• W4379649137 hasConcept C143409427 @default.
• W4379649137 hasConcept C204787440 @default.
• W4379649137 hasConcept C27081682 @default.
• W4379649137 hasConcept C2777575956 @default.
• W4379649137 hasConcept C2778216619 @default.
• W4379649137 hasConcept C2989005 @default.
• W4379649137 hasConcept C71924100 @default.
• W4379649137 hasConceptScore W4379649137C126322002 @default.
• W4379649137 hasConceptScore W4379649137C126838900 @default.
• W4379649137 hasConceptScore W4379649137C142724271 @default.
• W4379649137 hasConceptScore W4379649137C143409427 @default.
• W4379649137 hasConceptScore W4379649137C204787440 @default.
• W4379649137 hasConceptScore W4379649137C27081682 @default.
• W4379649137 hasConceptScore W4379649137C2777575956 @default.
• W4379649137 hasConceptScore W4379649137C2778216619 @default.
• W4379649137 hasConceptScore W4379649137C2989005 @default.
• W4379649137 hasConceptScore W4379649137C71924100 @default.
• W4379649137 hasIssue "Suppl 1" @default.
• W4379649137 hasLocation W43796491371 @default.
• W4379649137 hasOpenAccess W4379649137 @default.
• W4379649137 hasPrimaryLocation W43796491371 @default.
• W4379649137 hasRelatedWork W1937590427 @default.
• W4379649137 hasRelatedWork W2089714071 @default.
• W4379649137 hasRelatedWork W2156283049 @default.
• W4379649137 hasRelatedWork W2156609790 @default.
• W4379649137 hasRelatedWork W2316954211 @default.
• W4379649137 hasRelatedWork W2409398909 @default.
• W4379649137 hasRelatedWork W2427452099 @default.
• W4379649137 hasRelatedWork W2801007752 @default.
• W4379649137 hasRelatedWork W4245782456 @default.
• W4379649137 hasRelatedWork W78488825 @default.
• W4379649137 hasVolume "82" @default.
• W4379649137 isParatext "false" @default.
• W4379649137 isRetracted "false" @default.
• W4379649137 workType "article" @default.