Related Experiment Video
Updated: Jul 28, 2026
Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
Published on: April 13, 2013
Curved planar reformatting and convolutional neural network-based segmentation of the small bowel for visualization
Yechiel Lamash1, Sila Kurugol1, Moti Freiman1
1Computational Radiology Laboratory, Department of Radiology, Children's Hospital Boston, Harvard Medical School, Boston, Massachusetts, USA.
Background:
Contrast-enhanced MRI of the small bowel is an effective imaging sequence for the detection and characterization of disease burden in pediatric Crohn's disease (CD). However, visualization and quantification of disease burden requires scrolling back and forth through 3D images to follow the anatomy of the bowel, and it can be difficult to fully appreciate the extent of disease.
Purpose:
To develop and evaluate a method that offers better visualization and quantitative assessment of CD from MRI.
Study Type:
Retrospective.
Population:
Twenty-three pediatric patients with CD.
Field Strength/Sequence:
1.5T MRI system and T1 -weighted postcontrast VIBE sequence.
Assessment:
The convolutional neural network (CNN) segmentation of the bowel's lumen, wall, and background was compared with manual boundary delineation. We assessed the reproducibility and the capability of the extracted markers to differentiate between different levels of disease defined after a consensus review by two experienced radiologists.
Statistical Tests:
The segmentation algorithm was assessed using the Dice similarity coefficient (DSC) and boundary distances between the CNN and manual boundary delineations. The capability of the extracted markers to differentiate between different disease levels was determined using a t-test. The reproducibility of the extracted markers was assessed using the mean relative difference (MRD), Pearson correlation, and Bland-Altman analysis.
Results:
Our CNN exhibited DSCs of 75 ± 18%, 81 ± 8%, and 97 ± 2% for the lumen, wall, and background, respectively. The extracted markers of wall thickness at the location of min radius (P = 0.0013) and the median value of relative contrast enhancement (P = 0.0033) could differentiate active and nonactive disease segments. Other extracted markers could differentiate between segments with strictures and segments without strictures (P < 0.05). The observers' agreement in measuring stricture length was >3 times superior when computed on curved planar reformatting images compared with the conventional scheme.
Data Conclusion:
The results of this study show that the newly developed method is efficient for visualization and assessment of CD.
Level Of Evidence:
4 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2019;49:1565-1576.
Insights
A new convolutional neural network (CNN) method improves MRI visualization and quantification of pediatric Crohn's disease (CD) burden. This AI-driven approach enhances assessment accuracy for better patient management.
Area of Science:
- Medical Imaging
- Artificial Intelligence in Medicine
- Gastroenterology
Background:
- Contrast-enhanced MRI is crucial for assessing pediatric Crohn's disease (CD) burden.
- Current 3D MRI visualization requires extensive manual review, hindering comprehensive disease assessment.
- Improved methods are needed for accurate and efficient CD evaluation in children.
Purpose of the Study:
- To develop and validate an advanced method for enhanced visualization and quantitative assessment of CD using MRI.
- To leverage artificial intelligence for more precise disease burden evaluation.
- To improve the diagnostic capabilities of MRI in pediatric CD.
Main Methods:
- A retrospective study involving 23 pediatric patients with CD.
- Utilized a convolutional neural network (CNN) for segmenting bowel lumen, wall, and background from 1.5T MRI scans.
- Compared CNN segmentation with manual delineation and assessed marker reproducibility and disease differentiation capabilities.
Main Results:
- CNN achieved high segmentation accuracy (DSCs: 75-97%) compared to manual methods.
- Extracted MRI markers, including wall thickness and contrast enhancement, effectively differentiated active from non-active disease segments (P < 0.005).
- The method also distinguished between strictured and non-strictured segments (P < 0.05) and improved observer agreement for stricture length measurement.
Conclusions:
- The developed CNN-based method offers efficient and accurate visualization and quantitative assessment of CD from MRI.
- This AI approach shows significant potential for improving the evaluation of disease burden in pediatric Crohn's disease.
- The findings suggest a promising advancement in applying AI to medical imaging for pediatric gastroenterology.
