MRI-based Synthetic CT in the Detection of Structural Lesions in Patients with Suspected Sacroiliitis: Comparison with MRI
Bibliographic record
Abstract
Background Evaluation of structural lesions in the sacroiliac (SI) joints can improve the accuracy for diagnosis of spondyloarthritis. However, structural lesions, such as erosions, are difficult to assess on routine T1-weighted MRI scans. Purpose To determine the diagnostic performance of MRI-based synthetic CT (sCT) in the depiction of erosions, sclerosis, and ankylosis of the SI joints compared with T1-weighted MRI, with CT as the reference standard. Materials and Methods A prospective study (clinical trial registration no. B670201837885) was performed from February 2019 to November 2019. Adults were referred from a tertiary hospital rheumatology outpatient clinic with clinical suspicion of inflammatory sacroiliitis. MRI and CT of the SI joints were performed on the same day. SCT images were generated from MRI scans using a commercially available deep learning–based image synthesis method. Two readers independently recorded if structural lesions (erosions, sclerosis, and ankylosis) were present on T1-weighted MRI, sCT, and CT scans in different reading sessions, with readers blinded to clinical information and other images. Diagnostic performance of sCT and T1-weighted MRI scans were analyzed using generalized estimating equation models, with consensus results of CT as the reference standard. Results Thirty participants were included (16 men, 14 women; mean age, 40 years ± 10 [standard deviation]). Diagnostic accuracy of sCT was higher than that of T1-weighted MRI for erosion (94% vs 86%, P = .003), sclerosis (97% vs 81%, P < .001), and ankylosis (92% vs 84%, P = .04). With sCT, specificity for erosion detection (96% [95% CI: 90, 98] vs 89% [95% CI: 81, 94], P = .01] and sensitivity for detection of sclerosis [94% [95% CI: 87, 97] vs 20% [95% CI: 10, 35], P < .001] and ankylosis (93% [95% CI: 78, 98] vs 70% [95% CI: 47, 87], P = .001) were improved. Conclusion With CT as the reference standard, synthetic CT of the sacroiliac joints has better diagnostic performance in the detection of structural lesions in individuals suspected of having sacroiliitis compared with routine T1-weighted MRI. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by Fritz in this issue.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 teacher head, 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".