POS0989 DEVELOPMENT OF INTERNATIONAL CONSENSUS ON A STANDARDIZED IMAGE ACQUISITION PROTOCOL FOR DIAGNOSTIC EVALUATION OF THE SACROILIAC JOINTS BY MRI – AN ASAS-SPARTAN COLLABORATION.
Bibliographic record
Abstract
<h3>Background</h3> In 2009, ASAS published a ‘Definition of active sacroiliitis on MRI for classification of axial spondyloarthritis (axSpA)’. This definition relied on two MRI sequences to make this determination – semicoronal T1 and STIR. Since then, this approach has frequently been used for diagnosis, even though that was never the intent of the definition. In 2015, the European Society of Skeletal Radiology (ESSR) published its recommendations for an SIJ MRI image acquisition protocol (IAP) for diagnostic purposes that required 4 MRI sequences but there is still no IAP that has been widely accepted as a minimum standard worldwide. In 2020, an informal survey of 24 academic sites (12 Europe, 12 North America) confirmed that 24/24 sites performed a minimum of 3 MRI sequences for diagnosis (19 performed 4-8 sequences) because the 2-sequence protocol was considered inadequate. <h3>Objectives</h3> To develop the minimum requirements for a standardized IAP for MRI of the sacroiliac joints for diagnostic ascertainment of sacroiliitis. <h3>Methods</h3> All radiologist members of the ASAS and SPARTAN Classification in axSpA (CLASSIC) project, along with one European and one North American rheumatologist with extensive MRI experience in SpA clinical practice and research, were invited to participate in a consensus exercise. A draft IAP was circulated to all participants along with background information and justification for the draft proposal. Feedback on all issues was received by email, tabulated and recirculated. Participants were broadly in favour of the proposal and two months later a teleconference meeting took place and remaining points of contention were resolved. Examples of the proposed IAP performed on new, 10 and 22 years’ old MRI scanners were made available for review in DICOM format. Next the revised draft of the IAP was presented at the ASAS annual meeting to the entire membership on 14 January 2022, and voted on. <h3>Results</h3> A 4-sequence IAP, 3-semicoronal and 1-semiaxial, is recommended for diagnostic ascertainment of sacroiliitis and its differential diagnoses (Table 1). It must meet the following requirements: Semicoronal sequences should be parallel to the dorsal cortex of the S2 vertebral body, and include: 1) a sequence sensitive for the detection of active inflammation being T2-weighted with suppression of fat signal; 2) a sequence sensitive for the detection of structural damage in bone and bone marrow with T1-weighting; 3) a sequence that is designed to optimally depict the bone-cartilage interface of the articular surface and be sensitive for detection of bone erosion; plus 4) a semiaxial sequence sensitive for inflammation detection. The IAP was approved at the ASAS annual meeting by a vote of the entire membership with 91% in favour. <h3>Conclusion</h3> A standardized IAP for MRI of the sacroiliac joints for diagnostic ascertainment of sacroiliitis is recommended and should be comprised of a minimum of 4 sequences, in 2-planes, that will optimally visualize inflammation, structural damage, and the bone-cartilage interface. <h3>Disclosure of Interests</h3> Robert Lambert Paid instructor for: Novartis, Consultant of: Calyx, CARE Arthritis, Image Analysis Group, Xenofon Baraliakos Speakers bureau: Abbvie, Pfizer, MSD, UCB, Novartis, Lilly, Galapagos, Hexal, Paid instructor for: Abbvie, Pfizer, MSD, UCB, Novartis, Lilly, Galapagos, Hexal, Consultant of: Abbvie, Pfizer, MSD, UCB, Novartis, Lilly, Galapagos, Hexal, Grant/research support from: Abbvie, MSD, Novartis, Lilly, Stephanie Bernard Consultant of: Elsevier Amirsys, John Carrino Consultant of: Pfizer, Regeneron, Globus, Carestream, Image Analysis Group, Image Biopsy Lab, Torsten Diekhoff Speakers bureau: Novartis, MSD, Canon MS, Consultant of: Eli Lilly, Iris Eshed: None declared, Kay-Geert Hermann Speakers bureau: AbbVie, Pfizer, MSD, Novartis. Co-founder: BerlinFlame GmbH, Nele Herregods: None declared, Jacob L Jaremko: None declared, Lennart Jans: None declared, Anne Grethe Jurik: None declared, John O’Neill: None declared, Monique Reijnierse: None declared, Michael Tuite Consultant of: GE HealthCare, Walter P Maksymowych Speakers bureau: Abbvie, Janssen, Novartis, Pfizer, UCB, Consultant of: Abbvie, Boehringer Ingelheim, Celgene, Eli-Lilly, Galapagos, Novartis, Pfizer, UCB, Grant/research support from: Abbvie, Novartis, Pfizer, UCB
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.147 | 0.119 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.006 | 0.004 |
| Science and technology studies | 0.002 | 0.002 |
| Scholarly communication | 0.004 | 0.003 |
| Open science | 0.006 | 0.008 |
| Research integrity | 0.005 | 0.006 |
| Insufficient payload (model declined to judge) | 0.020 | 0.019 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".