MR imaging of the small bowel in Crohn disease

Machine learning methods in automated detection of CT enterography findings in Crohn's disease: A feasibility study

PURPOSE

Qualitative findings in Crohn's disease (CD) can be challenging to reliably report and quantify. We evaluated machine learning methodologies to both standardize the detection of common qualitative findings of ileal CD and determine finding spatial localization on CT enterography (CTE).

MATERIALS AND METHODS

Subjects with ileal CD and a CTE from a single center retrospective study between 2016 and 2021 were included. 165 CTEs were reviewed by two fellowship-trained abdominal radiologists for the presence and spatial distribution of five qualitative CD findings: mural enhancement, mural stratification, stenosis, wall thickening, and mesenteric fat stranding. A Random Forest (RF) ensemble model using automatically extracted specialist-directed bowel features and an unbiased convolutional neural network (CNN) were developed to predict the presence of qualitative findings. Model performance was assessed using area under the curve (AUC), sensitivity, specificity, accuracy, and kappa agreement statistics.

RESULTS

In 165 subjects with 29,895 individual qualitative finding assessments, agreement between radiologists for localization was good to very good (κ = 0.66 to 0.73), except for mesenteric fat stranding (κ = 0.47). RF prediction models had excellent performance, with an overall AUC, sensitivity, specificity of 0.91, 0.81 and 0.85, respectively. RF model and radiologist agreement for localization of CD findings approximated agreement between radiologists (κ = 0.67 to 0.76). Unbiased CNN models without benefit of disease knowledge had very similar performance to RF models which used specialist-defined imaging features.

CONCLUSION

Machine learning techniques for CTE image analysis can identify the presence, location, and distribution of qualitative CD findings with similar performance to experienced radiologists.

Artificial Intelligence for Disease Assessment in Inflammatory Bowel Disease: How Will it Change Our Practice?

Artificial intelligence (AI) has arrived and it will directly impact how we assess, monitor, and manage inflammatory bowel disease (IBD). Advances in the machine learning methodologies that power AI have produced astounding results for replicating expert judgment and predicting clinical outcomes, particularly in the analysis of imaging. This review will cover general concepts for AI in IBD, with descriptions of common machine learning methods, including decision trees and neural networks. Applications of AI in IBD will cover recent achievements in endoscopic image interpretation and scoring, new capabilities for cross-sectional image analysis, natural language processing for automated understanding of clinical text, and progress in AI-powered clinical decision support tools. In addition to detailing current evidence supporting the capabilities of AI for replicating expert clinical judgment, speculative commentary on how AI may advance concepts of disease activity assessment, care pathways, and pathophysiologic mechanisms of IBD will be addressed.

MRE Evaluation of Intestinal Inflammation: Qualitative and Quantitative Assessment

ABSTRACT

Magnetic resonance enterography (MRE) is a powerful tool for evaluation and management of patients with Crohn disease. Changes of active inflammation of the small bowel can reliably and reproducibly be detected and monitored. Findings indicative of active inflammation include bowel wall thickening, intramural edema and mural hyperenhancement. These changes are most commonly reported qualitatively; however, quantitative indices have also been developed and validated to measure and monitor inflammation both for clinical care and research purposes. This article describes the essential findings of active inflammation on MRE as well as the fundamentals of both qualitative and quantitative assessment and reporting.

Can ADC be used as a surrogate marker of response to therapy in intestinal tuberculosis?

PURPOSE

To evaluate the utility of Diffusion-weighted imaging (DWI) and apparent diffusion coefficient (ADC) in assessing treatment response in patients of intestinal tuberculosis (ITB).

METHOD AND MATERIALS

MR Enterography (MRE) was done for patients with suspicion of ITB and 19 patients with pre- and post-treatment imaging were included in the analysis. MRE included T1W, T2W, post-contrast T1W, and DWI sequences. DWI was done using b values-0, 400 and 800 s/mm², and ADC maps were generated. The trace DW images and ADC values were compared before and after therapy. Composite gold standard (clinical, colonoscopic criteria, and biopsy) was used to assess treatment response and to classify into no response, partial response, and complete response.

RESULTS

Thirty-one bowel segments were evaluated at baseline and after treatment in 19 patients. Prior to therapy, restricted diffusion was seen in 29/31 (93.5%) segments. After treatment, patients with either complete or partial response (27/31 segments, 15 patients) showed significant rise in mean ADC values from 1.1 ± 0.37 × 10^-3 to 2.1 ± 0.64 × 10^-3 mm²/s (p value < 0.05), whereas no significant change was found in mean ADC values of non-responders (4/29 segments in 4 patients) which increased from 1.0 ± 0.1 × 10^-3 mm²/s on baseline scan to 1.32 ± 0.2 × 10^-3 mm²/s on post-treatment scan (p value = 0.318). An increase in ADC value was found to be a reliable and objective marker of improvement with response to therapy.

CONCLUSION

ADC values show good correlation with treatment response in ITB and can be used for objectively quantifying it.

Lower Gastrointestinal Tract Applications of PET/Computed Tomography and PET/MR Imaging

This article discusses the role of PET/CT and PET/MR imaging in the evaluation of inflammatory and malignant disorders of the lower gastrointestinal tract. This includes a review of the current literature and a discussion of new and emerging research.

A Fixed Stricture on Routine Cross-sectional Imaging Predicts Disease-Related Complications and Adverse Outcomes in Patients with Crohn's Disease

BACKGROUND

Patients with Crohn's disease (CD) typically undergo multiple cross-sectional imaging exams including computed tomography and magnetic resonance enterography during the course of their disease. The aim was to identify imaging findings that predict future disease-related poor outcomes.

METHODS

This was a retrospective, case control study at a single tertiary center. Cases were CD patients diagnosed with complications (bowel obstruction, perforation, internal fistula, or abscess); controls were CD patients without complications. Two radiologists blinded to clinical outcomes, independently scored cross-sectional imaging examinations obtained before the complication.

RESULTS

One hundred eight patients (67 F; 41 M) with CD (51 cases; 57 controls) were included. For the cases, 21 had internal fistulae, 15 had bowel obstructions, 13 had abdominal abscesses, and 2 developed bowel perforations. Patients with complications were more likely to have a fixed small bowel stricture on cross-sectional imaging (P = 0.01). A patient with a stricture and upstream dilatation was 3.4 times more likely to develop a complication in the next 2 years. When present in the setting of hypervascularity and/or evidence of active inflammation, the risk increased further to 15-fold. Cases were more likely to be active smokers (29% versus 12%, P = 0.033). Cases had more evidence of inflammation based on higher Harvey Bradshaw Index values and inflammatory biomarkers and lower hemoglobin values.

CONCLUSIONS

Information from radiologic studies, especially the presence of fixed strictures, can predict future CD complications. These findings, along with smoking and ongoing inflammation, should alert the clinician to the possibility of future complications.

Nuclear Medicine Imaging in Pediatric Infection or Chronic Inflammatory Diseases

In this review article, we focus on the most recent applications of nuclear medicine techniques (mainly ^99mTc/¹¹¹In white blood cells (WBC) scan, [¹⁸F]-FDG-PET/CT, [¹⁸F]-FDG-PET/MRI, and ^99mTc-IL-2 scintigraphy) in the study of children affected by peripheral bone osteomyelitis, fungal infections, inflammatory bowel diseases, and type 1 diabetes, owing to recent important published evidences of their role in the management of these diseases. For osteomyelitis in children, both bone scintigraphy and [¹⁸F]-FDG-PET have a major advantage of assessing the whole body in one imaging session to confirm or exclude multifocal involvement, whereas WBC scan has a limited role. In children with fungal infections, [¹⁸F]-FDG-PET can help in defining the best location for biopsy and can help in evaluating the extent of the infection and organs involved (also sites that were not yet clinically apparent), although its main role is for therapy monitoring. In inflammatory bowel diseases, and Crohn disease in particular, WBC scan has been successfully used for many years, but it is now used only in case of doubtful magnetic resonance (MR) or when MR cannot be performed and endoscopy is inconclusive. By contrast, there is an accumulating evidence of the role of [¹⁸F]-FDG-PET in management of children with Crohn disease, and PET/MR could be a versatile and innovative hybrid imaging technique that combines the metabolic information of PET with the high soft tissue resolution of MR, particularly for distinguishing fibrotic from active strictures. Finally, there are several new radiopharmaceuticals that specifically target inflammatory cells involved in the pathogenesis of insulitis aiming at developing new specific immunotherapies and to select children candidates to these treatments for improving their quality of life.

Current tecniques and new perpectives research of magnetic resonance enterography in pediatric Crohn's disease

Crohn's disease affects more than 500000 individuals in the United States, and about 25% of cases are diagnosed during the pediatric period. Imaging of the bowel has undergone dramatic changes in the past two decades. The endoscopy with biopsy is generally considered the diagnostic reference standard, this combination can evaluates only the mucosa, not inflammation or fibrosis in the mucosa. Actually, the only modalities that can visualize submucosal tissues throughout the small bowel are the computed tomography (CT) enterography (CTE) with the magnetic resonance enterography (MRE). CT generally is highly utilized, but there is growing concern over ionizing radiation and cancer risk; it is a very important aspect to keep in consideration in pediatric patients. In contrast to CTE, MRE does not subject patients to ionizing radiation and can be used to detect detailed morphologic information and functional data of bowel disease, to monitor the effects of medical therapy more accurately, to detect residual active disease even in patients showing apparent clinical resolution and to guide treatment more accurately.

MRI for Crohn's Disease: Present and Future

Crohn's disease (CD) is a chronic inflammatory condition with relapsing-remitting behavior, often causing strictures or penetrating bowel damage. Its lifelong clinical course necessitates frequent assessment of disease activity and complications. Computed tomography (CT) enterography has been used as primary imaging modality; however, the concern for radiation hazard limits its use especially in younger population. Magnetic resonance (MR) imaging has advantages of avoiding radiation exposure, lower incidence of adverse events, ability to obtain dynamic information, and good soft-tissue resolution. MR enterography (MRE) with oral contrast agent has been used as primary MR imaging modality of CD with high sensitivity, specificity, and interobserver agreement. The extent of inflammation as well as transmural ulcers and fibrostenotic diseases can be detected with MRE. Novel MR techniques such as diffusion-weighted MRI (DWI), motility study, PET-MRI, and molecular imaging are currently investigated for further improvement of diagnosis and management of CD. MR spectroscopy is a remarkable molecular imaging tool to analyze metabolic profile of CD with human samples such as plasma, urine, or feces, as well as colonic mucosa itself.

MR imaging of distal ileal and colorectal chronic inflammatory bowel disease--diagnostic accuracy of 1.5 T and 3 T MRI compared to colonoscopy

PURPOSE

The aim of this study was to investigate the diagnostic accuracy of MR enterography (MRE) for detection of distal ileal and colorectal inflammatory bowel disease (IBD) and to evaluate whether 3 T MRI can provide a higher diagnostic performance compared to 1.5 T.

METHODS

A retrospective review of patients with known or suspected IBD who underwent MRE and colonoscopy within 3 months was performed. For analysis, the bowel was divided into six segments. Compared with colonoscopy, the accuracy values for MRI diagnosis of overall and each magnetic field strength were calculated, and the differences between 1.5 T and 3.0 T were compared. The image quality was scored separately for both field strengths and compared.

RESULTS

Eighty-eight patients were included in the study. On a patient basis, MRE had an overall sensitivity of 92.1 % and specificity of 72.0 %. On a segment basis, the sensitivity and specificity were 79.1 % and 93.6 %, respectively. Concerning severely inflamed segments, per-segment sensitivity increased from 79.1 to 94.7 %. The comparison of accuracy values between the two field strengths showed no statistically significant difference. B1 homogeneity and overall artifacts were not significantly different between 3.0 T and 1.5 T imaging. Compared to colonoscopy, MRI found four more fistulas confirmed at subsequent surgery.

CONCLUSIONS

MRI has a high diagnostic accuracy for detection of distal ileal and colorectal IBD. 3 T MRI can be considered equivalent but not superior compared to 1.5 T imaging in this context. In addition, our findings suggest MRE to be a valuable tool in detecting surgically relevant pathologies (fistulas) with higher accuracy than colonoscopy.

Comprehensive magnetic resonance enterography of Crohn's disease in the pediatric population: technique, interpretation, and management

Magnetic resonance enterography (MRE) plays a critical role in the management of Crohn's disease in the pediatric population. The ability to provide dynamic assessment of disease burden, complications, and therapeutic response without ionizing radiation makes it an ideal tool for younger patients requiring frequent follow-up. With a growing array of available treatment options, a sound understanding of MRE is critical in directing management aimed at curbing the physical and emotional morbidity associated with the lifelong condition. The goal of this article is to provide a practical overview of MRE in the pediatric population. This includes a review of our technique, approach to interpretation, pictorial collection of findings, and discussion of the role MRE plays in management.

Magnetic resonance imaging for follow-up of Crohn's disease

AIM: To investigate the value of magnetic resonance imaging (MRI) in the follow-up of Crohn's disease.

METHODS: Thirty three patients (including 24 men and 9 women) with Crohn's disease diagnosed from June 2007 to August 2013 were analyzed retrospectively. According to the clinical data, the patients were divided into either an active phase group or a chronic phase group. All the patients underwent MRI, and the images were reviewed for the mural thickness, the degree of stenosis of diseased bowel segments, mesenteric lymph nodes, and complications (abscess, inflammatory mass, and fistula). Disease activity was assessed according to a multi-parameter score, and MR findings were compared with clinical data.

RESULTS: Of all patients, 24 were clinically identified to have active disease and 9 to have chronic disease, and MR-E identified active disease in 21 cases and chronic disease in the remaining 12. Using clinical diagnosis as the "gold standard", the sensitivity, specificity, positive predictive value and negative predictive value of MR for the depiction of disease activity of CD were 79.2%, 77.8%, 90.5% and 58.3%, respectively. The diagnostic performance of MRI was well matched with that of clinical diagnosis (Kappa = 0.857, P < 0.05). In all the patients, 26 had two or more than 2 segments of the small bowl involved, 29 had the ileocecal area involved, 8 had enlarged mesareic lymph nodes, 17 had "comb sign", and 12 had fistula. There were significant differences in mural thickness, the degree of stenosis of diseased bowel segments, lymph nodes, "comb sign" and fistula between the active disease group and chronic disease group (P < 0.05 for all).

CONCLUSION: The diagnostic performance of MRE is well matched with clinical diagnosis. MRI is more effective in differentiating fistula and abscess, and more suitable for long-term follow-up.