126 - Optimizing MRI Imaging for Bitmap Printing Liver Models in Surgical Applications

Poster Board Number: 126
There are separate poster presentation times for odd and even posters.
Odd poster #s – first hour
Even poster #s – second hour

Abstract Body : Introduction/Objective: Hepatic surgeries are complex due to extensive vascularization caused by intertwining of branches of the hepatic arteries, portal veins, and biliary tree throughout the organ. This can lead to complications such as excessive bleeding during surgery or hemorrhaging post-surgery, thus presurgical planning requires cross-sectional imaging from multidetector computed tomography (CT) or magnetic resonance imaging (MRI). However, 2D sectional images require surgeons to mentally reconstruct the organ, adding cognitive load. Tangible 3D printed models of livers decrease surgical time, inpatient costs, and minimizes risks for patients. We will create a 3D printed liver model which directly translates vasculature from an image for presurgical planning and intraoperative guidance. Creating a methodology to augment the 2D images can create better resolution 3D printed models which accurately show vascular branching.

By increasing the efficiency of developing a 3D print liver model, the model can be regularly incorporated in surgical care plans for liver surgeries. Current methods of 3D printing use stereolithography (STL) file format requiring segmenting, editing, and co-registering structures prior to printing which is time-consuming and produces a crude representation of the structures. Bitmap printing, our novel method of printing, has been shown to produce more accurate models using less steps. However, bitmap printing requires a deeper understanding of source images, and few studies of 3D printed liver models are explicit in the settings of the source images used. Better imaging will improve voxel printing and provide a 3D model showing multigenerational branching that may give a better understanding of the patient-specific vasculature which will lead to reduced risk during surgery.

Methods: Anatomical structures are taken from several EOVIST triple-phase MRIs of the liver of a pediatric patient which will be edited individually and co-registered. Each scan varies in slice count, slice thickness, and resolution. Using the bitmap printing method, a colored histogram is applied directly to the voxelized volume created from the MRI image slices which will isolate the structures based on their intensity value. That stack of images will be fed to our patented software and printed on a multi-material polyjet printer.

Results: Using retrospective cases, several voxel-based liver models are printed using the methods above to understand differences in MRI image settings. All models will show hepatic arteries in red, portal vein in blue, and biliary tree in green in clear parenchyma. These models will provide a visual comparison of results from different settings to determine the best representation of hepatic anatomy that can be produced via bitmap printing.

Conclusion: Optimizing the 2D images leads to a high resolution and anatomically accurate 3D model of a patient’s liver via bitmap printing. This provides the ideal MRI settings to produce a 3D model for hepatic surgeries.

Significance/Implication: Bitmap printing reduces the barrier of incorporating 3D printing into surgical planning, thus potentially reducing surgical time, inpatient costs, and minimizing patient risk. By optimizing the 2D images, the benefits of bitmap printing can be amplified leading to higher resolution 3D printed models with more extensive branching. The completion of this study will allow for clinical trials using 3D printed models for hepatic surgeries.