ArticleEuropean radiology2026
Diagnostic interchangeability of deep-learning based Synth-STIR images generated from T1 and T2 weighted spine images.
Article in European radiology, 2026. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Cited by 2 papers.
What it found
Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.
The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.
The trial behind it
Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.
Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.
Who cites it
2 citing papers in PubMed.
- Seeing the Risk in Virtual Fat-Suppressed Spine MRI: External Benchmarking and Uncertainty-Guided Selective Risk Analysis.Journal of imaging informatics in medicine · 2026Article
- Intelligence Architectures and Machine Learning Applications in Contemporary Spine Care.Bioengineering (Basel, Switzerland) · 2025Review
Corrections and comments
PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.
Authors and funding
13 authors.
Funding
Abstract
objectivesTo evaluate image quality and diagnostic interchangeability of synth short-tau inversion recovery (STIR) generated by deep learning in comparison with standard STIR. MATERIALS AND
methodsThis prospective study recruited participants between July 2023 and August 2023. Participants were scanned with T1WI and T2WI, then generated Synth-STIR. Signal-to-noise ratios (SNR), contrast-to-noise ratios (CNR) were calculated for quantitative evaluation. Four independent, blinded radiologists performed subjective quality and lesion characteristic assessment. Wilcoxon tests were used to assess the differences in SNR, CNR, and subjective image quality. Various diagnostic findings pertinent to the spine were tested for interchangeability using the individual equivalence index (IEI). Inter-reader and intra-reader agreement and concordance were computed, and McNemar tests were performed for comprehensive evaluation.
resultsOne hundred ninety-nine participants (106 male patients, mean age 46.8 ± 16.9 years) were included. Compared to standard-STIR, Synth-STIR reduces sequence scanning time by approximately 180 s, has significantly higher SNR and CNR (p < 0.001). For artifacts, noise, sharpness, and diagnostic confidence, all readers agreed that Synth-STIR was significantly better than standard-STIR (all p < 0.001). In addition, the IEI was less than 1.61%. Kappa and Kendall showed a moderate to excellent agreement in the range of 0.52-0.97. There was no significant difference in the frequencies of the major features as reported with standard-STIR and Synth-STIR (p = 0.211-1).
conclusionSynth-STIR shows significantly higher SNR and CNR, and is diagnostically interchangeable with standard-STIR with a substantial overall reduction in the imaging time, thereby improving efficiency without sacrificing diagnostic value. KEY POINTS: Question Can generating STIR improve image quality while reducing spine MRI acquisition time in order to increase clinical spine MRI throughput? Findings With reduced acquisition time, Synth-STIR has significantly higher SNR and CNR than standard-STIR and can be interchangeably diagnosed with standard-STIR in detecting spinal abnormalities. Clinical relevance Our Synth-STIR provides the same high-quality images for clinical diagnosis as standard-STIR, while reducing scanning time for spine MRI protocols. Increase clinical spine MRI throughput.
Indexed as
Identifiers
40681776What OpenQuestion holds
Registered trials
Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the OpenQuestion graph.