RPS 801 - Advances in inflammatory bowel disease (IBD) imaging

Author Block: K. Johnson¹, I. De Kock², K. J. H. Bengtsson³, B. Janssens¹, A. Menys⁴, G. Bislenghi¹, C. W. P. Greer⁴, G. Bhatnagar⁴, B. Verstockt¹; ¹Leuven/BE, ²Ghent/BE, ³Lund/SE, ⁴London/UK
Purpose: The Magnetic Resonance Novel Index for Fistula Imaging in Crohn's Disease (MAGNIFI-CD) and 3D volumetry are emerging metrics for quantifying and visualizing perianal Crohn’s disease (pCD). The purpose of the study was to assess inter-reader agreement for both metrics in different pCD TOpClass categories among radiologists with varying levels of expertise in inflammatory bowel disease (IBD).
Methods or Background: Fifty pelvic MRIs from 50 unique pCD patients were retrospectively and randomly selected from a single-center IBD database: 25 represented minimal disease (suitable for repair; TOpClass 1–2a), while 25 represented disease requiring symptom control interventions/proctectomy (TOpClass 2b–3). Six radiologists independently assessed each MRI using a commercial MAGNIFI-CD module (Entrolytics by Motilent), blinded to clinical data. Three were IBD experts (>50 fistula MRI reads/year), three were non-IBD experts (<10/year). MAGNIFI-CD (0–25) was scored, and contrast-enhancing fistula volumes were manually segmented on T1 post-contrast axial images. Inter-reader agreement was analyzed through intraclass correlation coefficient (ICC).
Results or Findings: Inter-reader agreement for MAGNIFI-CD was ICC = 0.747 (95% CI: 0.640–0.835), among experts 0.786 (0.681–0.865) and non-experts 0.780 (0.610–0.877). Agreement for 3D volume was 0.887 (0.830–0.929), among experts 0.872 (0.792–0.923) and non-experts 0.893 (0.796–0.942). In TOpClass 1–2a, agreement for MAGNIFI-CD was 0.706 (0.561–0.834) and volume 0.537 (0.367–0.715). In TOpClass 2b–3, agreement was 0.664 (0.482–0.816) for MAGNIFI-CD and 0.885 (0.800–0.943) for volume.
Conclusion: MAGNIFI-CD and 3D fistula volume measurements showed substantial to almost perfect inter-reader agreement with comparable performance between experts and non-experts, underscoring their reliability for research and clinical use. Agreement for volume was, however, only moderate in mild disease (TOpClass 1–2a), indicating greater assessment difficulty in this subgroup.
Limitations: Single-center, small sub-groups.
Funding for this study: Swedish governmental funding of clinical research ALF, Maggie Stephens foundation, Gastroenterological Research Fund Sweden, the Royal Physiographic Society of Lund, Medical Society in Lund, Olle Olsson foundation, Nils-Magnus and Irma Ohlsson Foundation, Swedish Society of Radiology
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Local ethics committee approval (S69002)

Author Block: P. Minelli, A. Colarieti, B. Mattio, C. Garlisi, A. Carriero; Novara/IT
Purpose: To investigate the diagnostic performance of T2 mapping for the detection and quantification of
bowel wall oedema in patients with Crohn’s disease and to assess its correlation with fecal
calprotectin levels as a biochemical marker of intestinal inflammation.
Methods or Background: This retrospective, single-center study included 18 patients with histologically confirmed Crohn’s
disease who underwent magnetic resonance enterography at 3T. T2-mapping sequences were
acquired using a single-breath-hold axial gradient-echo protocol. Regions of interest (ROIs) were
delineated on thickened bowel segments previously identified on standard T2-weighted images.
Statistical analyses included Kolmogorov–Smirnov testing, paired Student’s t-tests, and Pearson
correlation analyses.
Results or Findings: Quantitative evaluation on non-contrast T2-weighted images revealed pathological mural thickening (>5 mm) in 15 of 18 patients, with a mean thickness of 9.5 ± 2.5 mm. T2 relaxation times followed a normal distribution in both normal (p = 0.21) and thickened (p = 0.11) bowel segments. Mean T2 values were significantly higher in thickened walls compared to normal segments (105.6 ± 4.3 ms vs. 51.8 ± 9.4 ms; p < 10⁻⁶). No significant correlation was found between T2 values and fecal calprotectin levels.
Conclusion: T2-mapping provides a quantitative alternative for evaluating intestinal inflammation in patients with contraindications to contrast agents, serving as a dependable tool for disease monitoring during follow-up.
Limitations: This study is limited by its monocentric, retrospective design and the relatively small patient cohort. Furthermore, as T2 mapping is a parametric sequence originally developed for cardiac magnetic resonance, its application to bowel imaging requires further validation in larger, multicenter studies.
Funding for this study: No funding was received for this study.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Approved by the local institutional ethics committee of AOU Maggiore della Carità, Novara, Italy (approval number: CE 090/2025).

Author Block: M. Zhang, Y. Wang, X. Shen, Z. Q. Zhu, R. Zhang, J. Lin, L. Huang, X. Li, C. Fan; Guangzhou/CN
Purpose: Current diagnostic tools for fibrosis assessment in Crohn’s disease (CD) remain suboptimal. MR elastography may advance toward addressing this limitation by offering tissue biomechanical characterization. Our aim was to evaluate MR elastography for intestinal fibrosis diagnosis using surgical histopathology and Piezo1 signaling as reference, and determine its incremental value over established MR enterography through machine-learning modelling.
Methods or Background: This prospective study analyzed 119 intestinal specimens from 45 CD participants undergoing preoperative MR elastography and enterography. Elastographic (shear-wave speed [SWS], loss angle [φ]), advanced (DWI, T1 mapping) and conventional enterographic
parameters were measured. Histopathological assessment included fibrosis scoring and collagen proportion analysis. Immunofluorescence Piezo1 served as stiffness reference standard. Three types of models (basic: conventional; extended: +advanced; complete: +elastographic) were developed using seven machine-learning algorithms, with performance assessed via AUC (DeLong’s test).
Results or Findings: SWS and φ significantly correlated with histological fibrosis score (r=0.642/0.408) and collagen proportion (r=0.666/0.461) (all P<0.01), with AUCs of 0.870 (95% CI: 0.803-0.937) for SWS and 0.734 (0.640-0.828) for φ in differentiating moderate-to-severe from none-to-mild fibrosis. Piezo1 level correlated with histological fibrosis score (r=0.750, P<0.001) and collagen proportion (r=0.859, P<0.001). Importantly, SWS also showed significant correlation with Piezo1 expression (r=0.669, P<0.05). XGBoost consistently outperformed other algorithms across all models. Complete modelXGBoost outperformed basic modelXGBoost (AUC, 0.962 vs. 0.799; DeLong’s test, P=0.029) and extended modelXGBoost (0.962 vs. 0.803; P=0.035) in validation set, confirming the added diagnostic value of MR elastography.
Conclusion: MR elastography emerges as a clinically valuable tool for quantifying intestinal fibrosis in CD, providing real-time biomechanical profiling that synergistically enhances MR enterography’s precision in fibrosis severity stratification.
Limitations: Single-center recruitment may limit population representativeness and spatial sampling differences between imaging slices and histological sections was challenging.
Funding for this study: This study was financially supported by Shenzhen Medical Research Fund (B2302008), Guangdong Basic and Applied Basic Research Foundation (2023B1515020070), National Natural Science Foundation of China (82270693 and 82070680), Fundamental Research Funds for the Central Universities, Sun Yat-sen University (24ykqb003).
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Institutional Review Board of The First Affiliated Hospital, Sun Yat-Sen University (Approval No. 2023488)

Author Block: B. Zeng¹, X. Wei², Z. Song³, Z. Huang⁴, Y. Liu¹; ¹Shenzheng/CN, ²Shenzhen/CN, ³Qingdao/CN, ⁴Guangzhou/CN
Purpose: To evaluate whether a diagnostic model combining conventional CT imaging features with quantitative parameters from dual-layer spectral detector CT (DLCT) improves the discrimination between active and remission phases of Crohn’s disease (CD).
Methods or Background: This retrospective study included 89 patients with endoscopically confirmed CD (51 active, 38 in remission) who underwent contrast-enhanced DLCT. Conventional imaging features (bowel wall thickening, enhancement pattern, strictures) and spectral parameters—including iodine concentration (IC), normalized IC, effective atomic number (Zeff), and 45 keV monoenergetic attenuation—were analyzed in both arterial and venous phases. Univariate and multivariate logistic regression were used to identify independent predictors of disease activity, which were then incorporated into a combined diagnostic model. Model performance was evaluated using receiver operating characteristic (ROC) curve analysis.
Results or Findings: Univariate analysis identified 24 significant variables. Multivariate analysis demonstrated that bowel wall thickness, stricture severity, enhancement pattern, and spectral parameters—particularly IC and Zeff—were independent predictors of active CD. The combined model exhibited excellent diagnostic performance, achieving an area under the curve (AUC) of 0.959 (95% CI: 0.911–0.985, p < 0.001), with a sensitivity of 92.2% and specificity of 86.1%.
Conclusion: The integration of DLCT-derived quantitative parameters with conventional CT features provides a highly accurate and noninvasive method for assessing Crohn’s disease activity, offering a robust tool for clinical disease monitoring and stratification.
Limitations: This study is limited by its single-center retrospective design and the sample size. Further validation in larger, prospective multicenter cohorts is warranted to confirm generalizability.
Funding for this study: Not Applicable
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Approved by Ethics Committee of Shenzhen Hospital of Southern Medical University

Author Block: K. A. Bhagwat; Harapanahalli/IN
Purpose: Learn different types of radiological reporting errors in MR Perianal Fistulography as per Kim and Mansfield classification
Methods or Background: MR Fistulography reporting needs adequate knowlegde of MR anatomy , acqusition , history , clinical examination to correctly report the study. However preliminary reports done by residents will have errors.
40 MRI Fistulography studies were marked on PACS , anonymized and were provided to 15 residents of radiology in third year of training to report.
Structured reporting format were provided and a line diagram were drawn for coronal and axial plane by residents.
Two consultant radiologists reviewed the reports done by residents as well as the reports done by consultants and tabulated the erros done by residents as compared to final reports. The consultants were blinded to reduce bias.
Results or Findings: All the reports done by 15 residents for 40 MR fistulograms( Total of 600 reports ) and report in PACS done by Consultants were reviewed for errors.

The errors were reviewed classified as per Kim and Mansfield classification ; into twelve types of errors.
Type 4 error, underreading was most common error found among the reports. Example : Tract from the second cutaneous opening was not identified.
Type 10 error, satisfaction of search , was the next common error . The secondary tract was being missed , once primary fistula tract is obvious.
Conclusion: Human error in interpretation of radiological images does occur for various reasons.. Understanding the errors done by residnets in training in reporting the MR fistulography will help in training them better.
Limitations: Since the residents were reporting the past studies done, they may not be in real time situation of doing a preliminary reporting or final report by the consultant.
Funding for this study: None.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:

Author Block: R. A. Pugliesi, M. Triscari Barberi, G. Roccella, R. Cannella, A. Mavaro, G. Lo Re; Palermo/IT
Purpose: To compare the diagnostic performance of non–contrast MRI sequences—balanced turbo/fast field echo (BTFE/BFFE), T2-weighted spectral attenuated inversion recovery (T2 SPAIR), and diffusion-weighted imaging (DWI)—with contrast-enhanced imaging for evaluating small-bowel disease activity in inflammatory bowel disease (IBD).
Methods or Background: This retrospective study included 207 adult patients who underwent magnetic resonance enterography: 145 with IBD and 62 without. Imaging features were recorded, and diagnostic performance of MRI sequences was assessed using the area under the curve (AUC), sensitivity, and specificity. Pairwise comparisons were performed with the DeLong test.
Results or Findings: IBD patients showed significantly higher prevalence of intestinal wall thickening (91.0% vs. 14.5%), ulcerations (34.5% vs. 0%), pseudopolyps (29.7% vs. 0%), vasa recta engorgement (66.9% vs. 1.6%), hypomobility (50.3% vs. 6.5%), adipose hypertrophy (68.3% vs. 3.2%), mesenteric lymphadenopathy (55.9% vs. 9.7%), and fistulas (14.5% vs. 1.6%) (all p < 0.01). Contrast-enhanced imaging achieved the highest diagnostic accuracy (AUC 0.860; sensitivity 80.0%; specificity 91.9%). Among non-contrast sequences, DWI (AUC 0.811; sensitivity 70.3%; specificity 91.9%) and BTFE/BFFE (AUC 0.800; sensitivity 64.8%; specificity 95.2%) performed significantly better than T2 SPAIR (AUC 0.670; sensitivity 46.9%; specificity 87.1%). Contrast-enhanced imaging was superior to all non-contrast sequences (p ≤ 0.036), while DWI and BTFE/BFFE showed no significant difference (p = 0.701).
Conclusion: Contrast-enhanced MRI demonstrated the highest diagnostic accuracy and should remain the reference standard in settings requiring maximal diagnostic confidence. However, DWI and BTFE/BFFE sequences showed high specificity, indicating their potential as reliable alternatives when contrast use is contraindicated or in follow-up scenarios requiring repeated imaging.
Limitations: Limitations include the retrospective design, potential selection bias
Funding for this study: n/a
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: study submitted and approved by the University of Palermo ethics committee

Author Block: C. Dennis, W. Weston, W. Blad, I. Naim, A. Menys, S. Taylor; London/UK
Purpose: Patients with ileo-anal pouches often develop intrusive increased bowel frequency. GLP-1 agonists reduce frequency possibly inhibiting motility. Using motility MRI (mMRI), we investigated the relationship between pouch motility, endoscopic inflammation and symptoms
Methods or Background: 30 ileoanal pouch patients (mean age 44 years, 9 female) and 10 controls (mean age 44, 5 male: 5 with non-colonic Crohns disease, 5 with ulcerative colitis) underwent MR enterography including a breath-hold balanced steady-state gradient echo motility sequence processed using GIQuant (Motilent) A radiologist placed regions of interest around the wall of the pouch and rectum (for controls), deriving a motility metric (expressed in arbitrary units (AU). Pouchoscopy was dichotomized into inflammation vs. no inflammation. Daily patient bowel frequency was dichotomized into =>10 and <10 , and symptoms into “highly 'symptomatic' vs 'coping. Motility was compared using Mann Whitney statistic.
Results or Findings: Five patients were excluded due to suboptimal pouch visualisation. Mean pouch motility was higher than rectal motility 157AU (25 to 391) vs. 59AU (23 to 104), p = 0.002. Patients with endoscopic inflammation (n=11) had lower motility than those without
(119AU vs 185AU, P = 0.05). Motility was higher in those with =>10 bowel movements (n=13) compared to <10 (205AU vs. 116AU, P = 0.007. Pouch motility was correlated to bowel movement frequency (Rho = 0.46, p =0.01), but not significantly different between highly symptomatic patients (n=9) vs. those coping (183AU vs. 132AU) P = 0.1.
Conclusion: mMRI quantified pouch motility is associated with bowel frequency and inflammation providing mechanistic insights for new therapeutic agents
Limitations: A retrospective study with a small sample size.
Funding for this study: n/a
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Data sharing ethics