David Lam1, Eric Yong2, Basil D'Souza1, Rodney Woods1. 1. Department of Colorectal Surgery, St. Vincent's Hospital Melbourne, Melbourne, Victoria, Australia. 2. Department of Radiology, St. Vincent's Hospital Melbourne, Melbourne, Victoria, Australia.
Abstract
BACKGROUND: Pelvic MRI allows for clear delineation of anatomy in Crohn's fistula-in-ano, although its interpretation is often difficult for nonradiologists. OBJECTIVE: The aim was to develop a 3-dimensional model where fistula tracts and their relationship to the sphincter complex can be accurately defined, which can then be rotated in multiple axes by the surgeon. DESIGN: A 3-dimensional model was created based on MRI images. An additional 3-dimensional T2-weighted sequence was added to the existing MRI protocol to obtain high-resolution images. Segmentation of the fistula tract and volume rendering of the segmented tract were performed to create the final model. SETTINGS: This was a single-center study conducted in Victoria, Australia. PATIENTS: All of the patients who had pelvic MRI for fistulating Crohn's disease between March 2016 and March 2017 had the additional MRI sequence. INTERVENTIONS: Postprocessing of MRI images was performed by a single radiologist. RESULTS: Total acquisition time for MRI images was extended to 31 minutes compared with the standard 2-dimensional protocol lasting 25 minutes. Additional postprocessing time used to create the model was ≈15 minutes. Two clinical vignettes using this model are presented and compared with conventional 2-dimensional MRI images to highlight the use of the 3-dimensional modeling technique. LIMITATIONS: This technique involves a semiautomatic process of fistula tract segmentation that requires radiologist expertise and additional postprocessing time. CONCLUSIONS: This 3-dimensional modeling technique enables accurate identification of tracts in Crohn's fistula-in-ano and improves spatial orientation for the surgeon. The model has the potential to be an invaluable preoperative tool to guide operative decision-making, as well as enabling the assessment of response to medical or surgical therapy.
BACKGROUND: Pelvic MRI allows for clear delineation of anatomy in Crohn's fistula-in-ano, although its interpretation is often difficult for nonradiologists. OBJECTIVE: The aim was to develop a 3-dimensional model where fistula tracts and their relationship to the sphincter complex can be accurately defined, which can then be rotated in multiple axes by the surgeon. DESIGN: A 3-dimensional model was created based on MRI images. An additional 3-dimensional T2-weighted sequence was added to the existing MRI protocol to obtain high-resolution images. Segmentation of the fistula tract and volume rendering of the segmented tract were performed to create the final model. SETTINGS: This was a single-center study conducted in Victoria, Australia. PATIENTS: All of the patients who had pelvic MRI for fistulating Crohn's disease between March 2016 and March 2017 had the additional MRI sequence. INTERVENTIONS: Postprocessing of MRI images was performed by a single radiologist. RESULTS: Total acquisition time for MRI images was extended to 31 minutes compared with the standard 2-dimensional protocol lasting 25 minutes. Additional postprocessing time used to create the model was ≈15 minutes. Two clinical vignettes using this model are presented and compared with conventional 2-dimensional MRI images to highlight the use of the 3-dimensional modeling technique. LIMITATIONS: This technique involves a semiautomatic process of fistula tract segmentation that requires radiologist expertise and additional postprocessing time. CONCLUSIONS: This 3-dimensional modeling technique enables accurate identification of tracts in Crohn's fistula-in-ano and improves spatial orientation for the surgeon. The model has the potential to be an invaluable preoperative tool to guide operative decision-making, as well as enabling the assessment of response to medical or surgical therapy.