POS1278 VALIDATION OF HANDHELD ULTRASOUND DEVICES FOR POINT OF CARE USE IN RHEUMATOLOGY: ANALYSIS FOR JOINTS AND NAILS
Bibliographic record
Abstract
Background: Ultrasonography (US) has experienced a rapid evolution in rheumatology. Despite many advantages being shown repeatedly, several barriers persist and stand in the way of a wider use of US in rheumatology, equipment cost being an important one. Hand-held US technology promises to take this cost down substantially. However, before it can be specifically used for rheumatology, its performance needs to be validated against gold-standard devices for key interventions. Objectives: We aim to test the concurrent validity of a handheld US device versus a gold-standard device to detect characteristic features of healthy and rheumatic joints (i.e., anatomical structures and vascular flow). Methods: Adult patients with peripheral PsA presenting with at least one tender and swollen joint were included. Each patient had consecutive US examinations using a handheld (Clarius Mobile Health Inc, HD3 L20 and L15 scanners) and a gold standard US device (GE LogicE9- E10) for detecting synovitis, erosions and nail disease. B mode and power Doppler images were saved for each site and lesion. Every image was given a unique identifier number at the end. A random order slide show was conducted for scoring, irrespective of the machine used or the anatomical site or patient assessed, to ensure blindness. Scoring was done using previously validated methods. Results: Thirty patients were recruited (n of scanned sites for synovitis=720; erosions=120; and nail disease=60). Patients had a mean±SD age of 56.6±11.2, with a median (IQR) disease duration of 7(2-12.25) years. 50% were female. On the day of the US, the median (IQR) tender joint count was 6(2-6) and swollen joint count was 2(1-2). The agreement between the handheld and gold standard devices for all elementary lesions assessed is shown in the table (Table 1). For detecting synovitis, the L15 and L20, had a kappa of 0.488 and 0.455 compared to the GE machine, respectively, with absolute agreement rates of 76.4-73.5%. For detecting the intrasynovial Doppler signals, the L15 had moderate agreement (kappa: 0.420, absolute agreement: 90.3%); and L20 had fair agreement (kappa: 0.367, absolute agreement: 90.9%). For erosions, there was moderate agreement with the L15 (kappa: 0.570, absolute agreement: 82.7%) and substantial agreement with L20 (kappa: 0.619, absolute agreement: 85.6%). The nail lesions were compared with L20, which showed a substantial agreement to detect the loss of trilaminar appearance (kappa: 0.643, absolute agreement: %87.9) and substantial agreement to detect nail bed vascularity (kappa: 0.663, absolute agreement: 89.5%). Conclusion: In this analysis, the handheld US devices show moderate-substantial agreement to detect synovitis, nail lesions and erosions. In Doppler activity, fair agreement was detected in the L20 device, while moderate agreement was detected in L15 device. These results encourage the use of handheld US devices for wider use. REFERENCES: NIL . Table 1The frequency, kappa and percent absolute agreement of elementary lesions on all three devices.Hand HeldL20 ScannerL15 ScannerGE LogicE9ErosionAbs-entPres-entKappa% Absolute agreementAbs-entPres-entKappa% Absolute agreementAbsent75110.61985.63340.57082.7Present520510Nail- loss of tri-laminar appearanceAbsent4230.64387.9Present49Nail DopplerAbsent820.66389.5Present443Synovitis-B ModeAbsent201510.45573.5154160.48876.4Present671265266Synovial DopplerAbsent384100.36790.924080.42090.3Present30141912 Acknowledgements: NIL . Disclosure of Interests: Seyyid Acikgoz: None declared, Ummugulsum Gazel: None declared, Suharsh Shah Full-time employee at Novartis Pharmaceuticals Canada Inc, Marie Maguin Full-time employee at Novartis Pharmaceuticals Canada Inc, Rohan Machhar Full-time employee at Novartis Pharmaceuticals Canada Inc, Patrick Leclerc Novartis employee, Lihi Eder Abbvie, Pfizer, UCB, Fresenius, Novartis, J&J, Abbvie, Pfizer, UCB, Novartis, Eli Lilly, BMS, Moonlake, J&J, Abbvie, Pfizer, UCB, Novartis, Eli Lilly, J&J, Fresenius Kabi, Gurjit Kaeley Research grants from Novartis, Gilead/ Galapagos, BMS,Janssen, Abbvie, Sibel Aydin received payment or honoraria for lectures, presentations, speaker's bureaus, manuscript writing, or educational events from Abbvie, Jannsen, Novartis, Pfizer, and UCB, received payment or honoraria for lectures, presentations, speaker's bureaus, manuscript writing, or educational events from Abbvie, Jannsen, Novartis, Pfizer, and UCB, stock options in Clarius, received consultant's fees from Abbvie, Celgene, Eli Lilly, Novartis, Pfizer, Sanofi, and UCB, received grants or contracts from Abbvie, Celgene, Eli Lilly, Jannsen, Novartis, Pfizer, Sanofi, and UCB. © The Authors 2025. This abstract is an open access article published in Annals of Rheumatic Diseases under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Neither EULAR nor the publisher make any representation as to the accuracy of the content. The authors are solely responsible for the content in their abstract including accuracy of the facts, statements, results, conclusion, citing resources etc.
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 | 0.009 | 0.015 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.007 | 0.003 |
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".