To date, ileocolonoscopy is the reference standard for evaluating disease activity in Crohn’s disease (CD). However, this is an invasive procedure that is, generally, poorly tolerated by the patient and which only allows for the mucosal surface to be visualized without assessing for intramural extent or extraluminal complications. Further complicating matters, many CD patients, especially children, have disease restricted to the small bowel, the majority of which is inaccessible by conventional endoscopy.
Magnetic Resonance Enterography (MRE) has emerged as an effective method for non-invasive imaging of the small bowel in patients with CD. MRE is especially useful for pediatric CD patients as doctors strive to spare children from the potentially harmful effects of radiation. Information conveying length of involvement, severity of inflammation, and luminal narrowing are required when assessing medical therapeutic response or deciding on the need for surgical treatment. However, due to the complex structure of the small bowel, visualization and quantification of disease burden requires scrolling back and forth through 2D images to follow the anatomy of the bowel which can make it difficult to fully appreciate the extent of disease.
The present technology is a software application (under development) enabling accurate assessment of Crohn’s disease (CD) extent from MRE data with automatic virtual unfolding of the small bowel. In a shift from current approaches, we are developing a deep-neural-networks (DNN) based automatic virtual unfolding algorithm that will enable accurate and reliable measurements of disease extent from MRE data to support the development of personalized treatment strategies for patients with CD.
- Time saving
- Consistent and objective results
Applications and Opportunities
- Automated image analysis solution
- Clinical decision support solution